Prevalence of occupational accidents and factors associated with deaths and disabilities in the shipbuilding industry: Comparisons of novice and skilled workers

Abstract

BACKGROUND:

The shipbuilding industry has various risks such as slipping, falling, mechanical, chemical, and confined space work.

OBJECTIVE:

The purpose of this study is to compare the characteristics of the occupational accidents between novice and skilled workers in the shipbuilding industry and to analyze factors affecting death and disability accidents.

METHODS:

From the national work-related compensation data of South Korea, an experimental design was established to analyze a population of occupational injuries related to workers working in the shipbuilding industry. This study compares accident characteristics of 2,069 injuries registered as work-related accidents. Also, logistic regression analysis is performed to identify the factors affecting death and disability accidents.

RESULTS:

The prevalence of occupational accidents caused by novices was high in older workers, female workers, foreign workers, irregular workers, or companies with less than 100 employees. The proportion of source of accident by novices was high in ‘manhole,’‘ladder,’ and ‘scaffold,’ while the prevalence of accident type was high in ‘fall,’ ‘cut/bruise,’ and ‘struck by.’ According to logistic regression analysis, if the face was injured, the possibility of being death or disability was higher than that of other parts of the body. Also, if the rupture caused the injury, the possibility of being death or disability was higher than the different types of injury.

CONCLUSIONS:

The results of this study are expected to be useful as basic data for the prevention of accidents of novice and death/disabilities in the shipbuilding industry.

Keywords

Shipyard work work experience work-related injury logistic regression analysis injury prevention

1 Introduction

Ship and boat building industry is concerned with the fabrication of a wide variety of marine vehicles [1]. The process of constructing ships includes assembling, installing, cleaning, painting, outfitting, and testing [2]. Shipbuilding workers perform a broad range of physical work and spend most of their time standing, walking, bending, and twisting parts of their bodies. They sometimes have to carry heavy equipment or materials. Shipbuilding works involve cutting, shaping, and assembling steel and other metals [3, 4]. Also, shipbuilding work is done in a variety of highly hazardous situations, such as confined spaces and considerable heights [4, 5]. Thus, there are many uncomfortable working postures, such as kneeling or overhead work [6, 7]. Also, a lot of work is performed outdoors, and the effects of weather extremes can cause or aggravate hazardous conditions. Additionally, some chemicals, paints, and solvents must be used, which may pose significant risks to workers [8, 9].

Shipyard work is traditionally hazardous, with an injury-accident rate more than twice that of construction and general industry [2]. Shipbuilding activities comprise some of the most hazardous workplace operations. Hazards include exposure to toxic substances, hazardous atmospheres, electrocution, falls, fires, and explosions [3 , 7]. Also, shipbuilding work is ergonomically hazardous, commonly requiring heavy physical work, uncomfortable postures, frequent bending, and twisting work [10, 11]. Most processes involve working with metal, and this typically produces noise levels above acceptable safe limits [2, 3].

Korea’s shipbuilding industry has developed into a powerful shipbuilding country but has a lot of problems in safety and health [4 , 8]. In the shipbuilding industry of South Korea in 2016, out of 229,899 workers working at 8,236 shipbuilding workplaces, 1,911 were approved by the Industrial Accident Compensation Act [12]. The accident rate of shipbuilding and repair work per 1,000 workers was 8.31, which was 1.347 times higher than that of 6.17 in the whole manufacturing industry [12]. In particular, due to the recession of the shipbuilding industry, in 2017, the number of workers in shipbuilding workshops decreased to 198,328. Still, the number of accidents approved by the Industrial Accident Compensation Act increased to 1,953 [13]. As a result, the accident rate per 1,000 persons accounted for 9.85, which was 2.035 times higher than that of the whole industry [13].

Previous research on the shipbuilding industry focused on accidental injuries and deaths [4 , 14]. Some investigations have emphasized that factors, such as age, work experience, work duration in a company, category of the worker, size of the company, type of accident, time of the accident, and weather, have influenced the severity of accidents [4, 14]. In the shipbuilding industry of South Korea, researches have been conducted on the risk factor analysis and ergonomics improvements [4 , 11], the implementation effect of ergonomics program [5, 15], and the risk assessment [10].

Identifying differences in incidental risk by work experience is of importance for producing suitable prevention policies. Efforts have been made to understand the characteristics of novice and skilled workers in various fields [16 –18]. However, there is a lack of research for the industrial accidents of novice workers in the shipbuilding industries. Therefore, this study aimed to compare the characteristics of occupational accidents between novices and skilled workers in the shipbuilding industry. Also, this study aimed to analyze the factors affecting death and disability accidents.

2 Methods

2.1 National work-related compensation data

In the Republic of Korea, employers have to offer workers’ compensation insurance under the Industrial Safety and Health Act. The compensation data for occupational injuries and illnesses [19] are compiled and maintained by the Ministry of Labor [16 , 20–22]. In this study, an experimental design was established to analyze a population of occupational accidents related to workers working in the shipbuilding industry, excluding occupational illness.

In the shipbuilding industry of the Republic of Korea, 2,252 workers were approved as work-related injuries in 2016 and 2017. This study focuses on occupational accidents occurring in the workplace except for accidents caused by commuting or outside activities. After classifying accidents happened in the workplace, 2,069 injuries were extracted from a total of 2,252 injuries.

2.2 Comparison of occupational accidents between novice and skilled worker

A novice is defined as a worker with less than one year of work experience. A skilled worker is defined as a worker with more than 10 years of work experience. From the national compensation dataset, we used the variables on worker-related factors and accident-related factors. This study compares the characteristics of worker-related factors and accident-related factors between novice and skilled workers in the shipbuilding industry. The variables of the injured workers are expressed by injury severity, gender, age, nationality, employment conditions, and company size. Accident-related variables are expressed in terms of source of accident, accident type, injured part of body, and injury type.

Among 2,069 injuries, 1,299 (62.8%) were injuries of less than one year of work experience, and 287 (13.9%) were injuries of more than 10 years. In this study, χ² test was performed at a significance level of 0.05 to test whether there is a difference in the distribution of injured persons between novice and skilled workers.

2.3 Logistic regression modeling on factors associated with deaths and disabilities

In this study, logistic regression analysis is performed to identify the factors affecting the deaths and disabilities of the shipbuilding industry. The logistic regression model describes the relationship between a binary dependent variable (1 = death or disability, 0 = otherwise) and independent variables (worker-related factors and accident-related factors). The independent variables were gender, age, nationality, type of contract, work experience, company size, accident type, source of accident, injury type, and injured part of body.

Table 1 shows the research variables used in the logistic regression analysis. For the logistic regression analysis, the block input method was used.

Table 1
Definition of research variables

Variable Description

Dependent variable

Injury severity 1, death or disability; 0, otherwise

Independent variable

Worker-related factors

Gender Male; Female

Age (years) < 40; 40–49; 50–59;≥60

Nationality Korean; Foreigner

Contract Regular; Temporary

Work experience (years) < 1; 1–9;≥10

Company size (employees) < 15; 15–99; 100–1999;≥2000

Accident-related factors

Accident type Slip; Struck by; Fall; Caught in; Explosion/burn; Cut/bruise

Source of accident Materials/parts/line; Tool/equipment; Floor/stairs; Scaffold;

Manhole; Ladder/heights; Crane; Other

Injury type Rupture; Fracture; Cut/bruise; Sprain; Burn; Other

Injured body part Head/face; Neck/shoulder; Trunk; Arm/hand; Leg/foot; Other

Variable	Description
Dependent variable
Injury severity	1, death or disability; 0, otherwise
Independent variable
Worker-related factors
Gender	Male; Female
Age (years)	< 40; 40–49; 50–59;≥60
Nationality	Korean; Foreigner
Contract	Regular; Temporary
Work experience (years)	< 1; 1–9;≥10
Company size (employees)	< 15; 15–99; 100–1999;≥2000
Accident-related factors
Accident type	Slip; Struck by; Fall; Caught in; Explosion/burn; Cut/bruise
Source of accident	Materials/parts/line; Tool/equipment; Floor/stairs; Scaffold;
	Manhole; Ladder/heights; Crane; Other
Injury type	Rupture; Fracture; Cut/bruise; Sprain; Burn; Other
Injured body part	Head/face; Neck/shoulder; Trunk; Arm/hand; Leg/foot; Other

2.4 Validity and reliability of this study

This study performed a reliability analysis to verify whether the independent variables for dependent variables were reliable. The ability of the model to discriminate between the two groups defined by the response variable is evaluated by correct classification table value. Also, the model is tested by the G value of the model coefficient. The goodness of fit of a model measures how well the model describes the response variable. The Hosmer–Lemeshow test is used for assessing the goodness of fit of a model [23]. The explanatory power of the dependent variable was assessed by the Nalgelkerke R² [23]. The SPSS version 18.0 statistical software package was used for all data analyses.

3 Results

3.1 Characteristics of injured workers by work experience

Table 2 shows the distribution of injured persons by work experience according to worker-related factors.

Table 2
Distribution of injured workers by work experience

Variables Work experience (years)

< 1 1–9 ≥10 Total

n % n % n % n %_row %_column

Severity Death 25 58.1 15 34.9 3 7.0 43 100.0 2.1

Disability 261 61.0 92 21.5 75 17.5 428 100.0 20.7

Injury 1013 63.4 376 23.5 209 13.1 1598 100.0 77.2

Gender Male 1190 61.9 450 23.4 281 14.6 1921 100.0 92.8

Female 109 73.6 33 22.3 6 4.1 148 100.0 7.2

Age < 40 335 57.0 214 36.4 39 6.6 588 100.0 28.4

40–49 357 61.9 115 19.9 105 18.2 577 100.0 27.9

50–59 377 65.8 92 16.1 104 18.2 573 100.0 27.7

≥60 230 69.5 62 18.7 39 11.8 331 100.0 16.0

Nationality Korean 1194 61.4 463 23.8 287 14.8 1944 100.0 94.0

Foreigner 105 84.0 20 16.0 0 0 125 100.0 6.0

Contract Regular 655 47.3 443 32.0 286 20.7 1384 100.0 66.9

Temporary 644 94.0 40 5.8 1 0.1 685 100.0 33.1

Company size < 15 652 86.6 91 12.1 10 1.3 753 100.0 36.4

15–99 445 76.1 133 22.7 7 1.2 585 100.0 28.3

100–1999 196 58.5 113 33.7 26 7.8 335 100.0 16.2

≥2000 6 1.5 146 36.9 244 61.6 396 100.0 19.1

Total 1299 62.8 483 23.3 287 13.9 2069 100.0 100.0

Variables		Work experience (years)
Severity	Death	25	58.1	15	34.9	3	7.0	43	100.0	2.1
	Disability	261	61.0	92	21.5	75	17.5	428	100.0	20.7
	Injury	1013	63.4	376	23.5	209	13.1	1598	100.0	77.2
Gender	Male	1190	61.9	450	23.4	281	14.6	1921	100.0	92.8
	Female	109	73.6	33	22.3	6	4.1	148	100.0	7.2
Age	< 40	335	57.0	214	36.4	39	6.6	588	100.0	28.4
	40–49	357	61.9	115	19.9	105	18.2	577	100.0	27.9
	50–59	377	65.8	92	16.1	104	18.2	573	100.0	27.7
	≥60	230	69.5	62	18.7	39	11.8	331	100.0	16.0
Nationality	Korean	1194	61.4	463	23.8	287	14.8	1944	100.0	94.0
	Foreigner	105	84.0	20	16.0	0	0	125	100.0	6.0
Contract	Regular	655	47.3	443	32.0	286	20.7	1384	100.0	66.9
	Temporary	644	94.0	40	5.8	1	0.1	685	100.0	33.1
Company size	< 15	652	86.6	91	12.1	10	1.3	753	100.0	36.4
	15–99	445	76.1	133	22.7	7	1.2	585	100.0	28.3
	100–1999	196	58.5	113	33.7	26	7.8	335	100.0	16.2
	≥2000	6	1.5	146	36.9	244	61.6	396	100.0	19.1
Total		1299	62.8	483	23.3	287	13.9	2069	100.0	100.0

Among 2,069 injuries, 43 (2.1%) were deaths, 428 (20.7%) were disabilities, and 1,598 (77.2%) were injuries. The occupational accident in the shipbuilding industry is as severe as 22.8%of the total injury is death or disability. There was a significant difference in the distribution of injured persons according to injury severity and to work experience had (χ ² = 9.878, p = 0.043). In the case of less than 1 year of work experience, injured persons (78.0%) have a higher proportion, while those who have more than 10 years have a higher percentage of disabled persons (26.1%).

Of the total injuries, 92.8%were men, and 7.2%were women. There was a difference in the distribution of injured persons according to gender and work experience (χ² = 14.147, p = 0.001). The percentage of female injured persons less than one year of experience (73.6%) was higher than that of men (61.9%). The ratio of male deaths and disabilities over 10 years of work experience (14.6%) was higher than that of female deaths and disabilities (4.1%).

Of the total injuries, 28.4%were under age 40, 27.9%were in their forties, 27.7%were in their 50s, and 16.0%were over 60 years old. There was a difference in the age distribution of injured persons according to work experience (χ² = 106.868, p < 0.001). As the age increase, the ratio of novice tends to increase. This means that older workers who joined the company after retirement have a high prevalence of accidents within one year of joining the company. Older workers are more likely to cause injuries in their workplaces such as darker, heights, heavy materials handling, and high-temperature work, because of their inability to adapt to workplaces.

Only 6.0%of the total disasters were foreigners. There were differences in the distribution of injured persons by work experience and nationality (x² = 31.066, p < 0.001). The proportion of foreigners with novice was 84.0%, which was much higher than that of Koreans (61.4%).

Of the total injuries, 66.9%were full-time regular workers, and 33.1%were non-regular workers. The distribution of occupational injuries by employment type and work experience was found to differ (x² = 432.578, p < 0.001). The ratio of novice to non-regular workers is very high at 94.0%, while that of regular workers is low at 47.3%, and that of skilled workers is relatively high at 20.7%.

Of the total injuries, 36.4%have occurred in the workplaces with less than 15 employees, and 64.7%have occurred in the workplaces with fewer than 100 employees. There was a difference in the distribution of injured persons by work experience and company size (x² = 1,222.579, p < 0.001). The percentage of the novice in the workplaces with less than 15 employees was high at 86.6%, but only 1.5%at the workplaces with 2000 workers or more. The larger the company, the lower the percentage of the novice. On the other hand, in the case of a workplace with 2000 employees, the proportion of skilled workers accounted for 61.6%. Therefore, education and training should be concentrated in small workplaces with fewer than 100 employees. Also, the work environment of these small-scale workplaces must be continuously improved.

3.2 Characteristics of occupational accidents by work experience

Table 3 shows the distribution of injured persons by work experience according to accident-related factors. Among the total injuries, the percentage of accidents caused by ‘struck by’ was the highest at 32.1%, followed by ‘fall’ (22.9%), ‘slip’ (21.4%), ‘caught in’ (14.3%), ‘explosion/burn’ (4.9%), and ‘cut/bruise’ (4.4%). There were differences in the distribution of injured persons by type of accidents according to work experience (x² = 48.768, p < 0.001). The percentage of ‘fall’ (71.9%) and ‘cut/bruise’ (70.7%) was particularly high in the novice. While, ‘caught in’ (20.0%), ‘slip’ (16.7%), ‘explosion/burn’ (14.7%), and ‘struck by’ (14.0%) were relatively high in the skilled workers.

Table 3
Distribution of occupational accidents by work experience

Variables Work experience (years)

< 1 1–9 ≥10 Total

n % n % n % n %_row %_column

Accident type Struck by 426 64.2 145 21.8 93 14.0 664 100.0 32.1

Fall 341 71.9 95 20.0 38 8.0 474 100.0 22.9

Slip 259 58.6 109 24.7 74 16.7 442 100.0 21.4

Caught in 152 51.5 84 28.5 59 20.0 295 100.0 14.3

Explosion/burn 56 54.9 31 30.4 15 14.7 102 100.0 4.9

Cut/bruise 65 70.7 19 20.7 8 8.7 92 100.0 4.4

Source of accident Materials/parts/line 366 60.7 132 21.9 105 17.4 603 100.0 29.1

Tool/equipment 224 60.7 95 25.7 50 13.6 369 100.0 17.8

Floor/stairs 160 58.0 68 24.6 48 17.4 276 100.0 13.3

Scaffold 150 68.2 47 21.4 23 10.5 220 100.0 10.6

Manhole 140 71.8 38 19.5 17 8.7 195 100.0 9.4

Ladder/heights 124 70.1 40 22.6 13 7.3 177 100.0 8.6

Crane 114 62.0 48 26.1 22 12.0 184 100.0 8.9

Other 21 46.7 15 33.3 9 20.0 45 100.0 2.2

Injury type Fracture 775 63.9 274 22.6 164 13.5 1213 100.0 58.6

Rupture 184 59.0 78 25.0 50 16.0 312 100.0 15.1

Cut/bruise 137 68.2 37 18.4 27 13.4 201 100.0 9.7

Sprain 84 60.0 34 24.3 22 15.7 140 100.0 6.8

Burn 49 52.7 30 32.3 14 15.1 93 100.0 4.5

Other 70 63.6 30 27.3 10 9.1 110 100.0 5.3

Injured part Head/face 135 64.9 51 24.5 22 10.6 208 100.0 10.1

Neck/shoulder 76 62.8 26 21.5 19 15.7 121 100.0 5.8

Trunk 237 66.2 81 22.6 40 11.2 358 100.0 17.3

Arm/hand 406 59.9 153 22.6 119 17.6 678 100.0 32.8

Leg/foot 425 63.5 158 23.6 86 12.9 669 100.0 32.3

Other 20 57.1 14 40.0 1 2.9 35 100.0 1.7

Total 1299 62.8 483 23.3 287 13.9 2069 100.0 100.0

Variables		Work experience (years)
Accident type	Struck by	426	64.2	145	21.8	93	14.0	664	100.0	32.1
	Fall	341	71.9	95	20.0	38	8.0	474	100.0	22.9
	Slip	259	58.6	109	24.7	74	16.7	442	100.0	21.4
	Caught in	152	51.5	84	28.5	59	20.0	295	100.0	14.3
	Explosion/burn	56	54.9	31	30.4	15	14.7	102	100.0	4.9
	Cut/bruise	65	70.7	19	20.7	8	8.7	92	100.0	4.4
Source of accident	Materials/parts/line	366	60.7	132	21.9	105	17.4	603	100.0	29.1
	Tool/equipment	224	60.7	95	25.7	50	13.6	369	100.0	17.8
	Floor/stairs	160	58.0	68	24.6	48	17.4	276	100.0	13.3
	Scaffold	150	68.2	47	21.4	23	10.5	220	100.0	10.6
	Manhole	140	71.8	38	19.5	17	8.7	195	100.0	9.4
	Ladder/heights	124	70.1	40	22.6	13	7.3	177	100.0	8.6
	Crane	114	62.0	48	26.1	22	12.0	184	100.0	8.9
	Other	21	46.7	15	33.3	9	20.0	45	100.0	2.2
Injury type	Fracture	775	63.9	274	22.6	164	13.5	1213	100.0	58.6
	Rupture	184	59.0	78	25.0	50	16.0	312	100.0	15.1
	Cut/bruise	137	68.2	37	18.4	27	13.4	201	100.0	9.7
	Sprain	84	60.0	34	24.3	22	15.7	140	100.0	6.8
	Burn	49	52.7	30	32.3	14	15.1	93	100.0	4.5
	Other	70	63.6	30	27.3	10	9.1	110	100.0	5.3
Injured part	Head/face	135	64.9	51	24.5	22	10.6	208	100.0	10.1
	Neck/shoulder	76	62.8	26	21.5	19	15.7	121	100.0	5.8
	Trunk	237	66.2	81	22.6	40	11.2	358	100.0	17.3
	Arm/hand	406	59.9	153	22.6	119	17.6	678	100.0	32.8
	Leg/foot	425	63.5	158	23.6	86	12.9	669	100.0	32.3
	Other	20	57.1	14	40.0	1	2.9	35	100.0	1.7
Total		1299	62.8	483	23.3	287	13.9	2069	100.0	100.0

Among the total injuries, ‘material/part/line’ occupied the highest percentage of accidents with 29.1%, followed by ‘tool/equipment’ (17.8%), ‘floor/stairs’ (13.3%), ‘scaffold’ (10.6%), ‘manhole’ (9.4%), ‘crane’ (8.9%), and ‘ladder (8.6%). There was a difference in the distribution of injured persons by the source of accident and work experience (χ² = 35.103, p = 0.001). The percentage of accidents caused by ‘manhole’ (71.8%), ‘ladder’ (70.1%), or ‘scaffold’ (68.2%) was higher in the novice. In comparison, the percentage of accidents caused by ‘material/part/line’ (17.4%), or ‘floor/stairs’ (17.4%) was relatively high in skilled workers.

‘Fracture’ accounted for 58.6%, followed by ‘rupture’ (15.1%), ‘cut/bruise’ (9.7%), ‘sprain’ (6.8%), and ‘burn’ (4.5%). There was no difference in the distribution of injured persons according to the type of injury and work experience (χ² = 28.612, p = 0.053).

The body site most vulnerable to injuries was arm/hand/finger (32.8%), followed by leg/feet (32.3%), trunk (17.3%), and head/face (10.1%). There was a difference in the distribution of injured persons according to the body part and work experience (x² = 20.676, p = 0.023). The portion of body site caused by novices was high in the trunk (66.2%), head/face (64.9%), or leg/foot (63.5%), while the portion caused by skilled workers was high in the arm/hand (17.6%) or neck/shoulder (15.7%).

3.3 Logistic regression analysis on deaths and disabilities

Table 4 shows the final results of the logistic regression analysis according to the deaths and disabilities (classification table value = 77.2%). The logistic regression model was tested by G value of the model coefficient (χ² = 763.969, p < 0.001). The explanatory power of the dependent variable was the Nagelkerke value (0.412), and the fitness test for the variable was Hosmer and Lemeshow (χ² = 7.252, significance value = 0.510).

Table 4
Results of logistic regress analysis on deaths and disabilities

Prevalence Crude OR CI p value

Variables n % ( )

Work experience 0.037^*

≥10 (ref) 287 27.2

< 1 1299 22.0 0.738 0.547 0.996 0.047^*

1–9 483 22.2 0.664 0.481 0.917 0.013^*

Nationality

Korean (ref) 1944 22.3

Foreigner 125 30.4 1.685 1.116 2.544 0.013^*

Contract

Regular (ref) 1384 24.4

Temporary 685 19.4 0.734 0.566 0.953 0.020^*

Accident type 0.002^*

Slip (ref) 442 19.5

Fall 474 24.5 1.297 0.941 1.789 0.112

Struck by 664 22.4 0.987 0.734 1.329 0.933

Caught in 295 30.8 1.162 0.796 1.696 0.438

Cut/bruise 92 18.5 0.494 0.270 0.906 0.023^*

Explosion/burn 102 11.8 7.158 1.828 28.038 0.005^*

Injury type < 0.001^*

Rupture (ref) 312 27.6

Sprain 140 5.7 0.147 0.068 0.314 <0.001^*

Fracture 1213 24.2 0.563 0.423 0.751 <0.001^*

Cut/bruise 201 22.9 0.471 0.310 0.716 <0.001^*

Burn 93 5.4 0.012 0.002 0.062 <0.001^*

Other 110 30.0 0.501 0.289 0.869 0.014^*

Injured part of body < 0.001^*

Head/face (ref) 208 19.7

Neck/shoulder 121 12.4 0.397 0.215 0.733 0.003^*

Trunk 358 19.8 0.552 0.373 0.815 0.003^*

Arm/hand 678 29.5 1.036 0.739 1.452 0.837

Leg/foot 669 19.7 0.530 0.390 0.720 <0.001^*

Multiple parts 35 34.3 1.831 0.788 4.254 0.160

Total 2069 22.8

			Prevalence	Crude OR	CI	p value
Work experience							0.037^*
	≥10 (ref)	287	27.2
	< 1	1299	22.0	0.738	0.547	0.996	0.047^*
	1–9	483	22.2	0.664	0.481	0.917	0.013^*
Nationality
	Korean (ref)	1944	22.3
	Foreigner	125	30.4	1.685	1.116	2.544	0.013^*
Contract
	Regular (ref)	1384	24.4
	Temporary	685	19.4	0.734	0.566	0.953	0.020^*
Accident type							0.002^*
	Slip (ref)	442	19.5
	Fall	474	24.5	1.297	0.941	1.789	0.112
	Struck by	664	22.4	0.987	0.734	1.329	0.933
	Caught in	295	30.8	1.162	0.796	1.696	0.438
	Cut/bruise	92	18.5	0.494	0.270	0.906	0.023^*
	Explosion/burn	102	11.8	7.158	1.828	28.038	0.005^*
Injury type							< 0.001^*
	Rupture (ref)	312	27.6
	Sprain	140	5.7	0.147	0.068	0.314	<0.001^*
	Fracture	1213	24.2	0.563	0.423	0.751	<0.001^*
	Cut/bruise	201	22.9	0.471	0.310	0.716	<0.001^*
	Burn	93	5.4	0.012	0.002	0.062	<0.001^*
	Other	110	30.0	0.501	0.289	0.869	0.014^*
Injured part of body							< 0.001^*
	Head/face (ref)	208	19.7
	Neck/shoulder	121	12.4	0.397	0.215	0.733	0.003^*
	Trunk	358	19.8	0.552	0.373	0.815	0.003^*
	Arm/hand	678	29.5	1.036	0.739	1.452	0.837
	Leg/foot	669	19.7	0.530	0.390	0.720	<0.001^*
	Multiple parts	35	34.3	1.831	0.788	4.254	0.160
Total		2069	22.8

^*p < 0.05.

In Table 4, skilled workers showed that the possibility of being death or disability due to an accident was higher than that of a novice (OR = 0.738, p = 0.047) or 1–9 years of work experience (OR = 0.664, p = 0.013). Foreign workers were 1.685 times more likely to be death or disability due to accidents than Korean workers (OR = 1.685, p = 0.013). Also, regular workers are more likely to be death or disabilities due to accidents than temporary workers (OR = 0.734, p = 0.020).

Among the types of accidents, it was found that an ‘explosion/burn’ accident is 7.158 times more likely to be a death or a disability than an accident caused by slip, while a ‘cut/bruise’ accident is less than an accident caused by slip,

In Table 4, rupture injuries were more likely to be death or disabilities than fractures (OR = 0.563, p < 0.001), cut/bruise (OR = 0.471, p < 0.001), sprain (OR = 0.147, p < 0.001), and burn (OR = 0.012, p < 0.001). In the injured part of body, head/facial part were more likely to cause death or disability than trunk (OR = 0.552, p = 0.003), leg/foot (OR = 0.530, p < 0.001), and neck/shoulder (OR = 0.397, p = 0.003).

4 Discussion

In this study, there are differences in the characteristics of occupational accidents between novice and skilled workers.

According to the results of the study, 62.8%of the total approved injuries were novices with less than one year of work experience. It suggests that education and training are necessary to get used to the work environment after entering the company. The proportion of novices tended to increase with age, and older workers with over 60 years old were 69.5%. With the growing ranks of the elderly population, the important issue is to provide sufficient opportunities for re-employment, continuing education, and work environment [24]. There is a need to design workplaces that accommodate the characteristics of older novice workers. The proportion of novices was higher in foreign workers than that of Korean workers. Also, the proportion of novices was higher in temporary employees than that of regular employees. Kim and Jeong [20] pointed out that universal safety ensures a certain level of health and safety for all workers regardless of their nationality as well as their contract statuses. Shipbuilding workers working on a small company may be more likely to be poor workplace and lacking a protective structure [4]. The proportion of novice injured workers was high in small businesses with fewer than 15 employees.

The percentage of type of accident caused by novices was high in ‘fall,’ ‘cut/bruise,’ or ‘struck by.’ On the other hand, the portion of source of the accident caused by novices was high in ‘manhole, ‘ladder, or ‘scaffold. Shipyard work includes exposure to toxic substances, hazardous atmospheres, dark workplace, falls, fires, and explosions [3, 9]. It is difficult for novice workers to recognize hazards when the work environment changes. Therefore, it is necessary to educate and train novice workers to become familiar with the workplace.

Occupational accidents in the shipbuilding industry are relatively high in the proportion of deaths and disabilities. In this study, the percentage of injured with disabilities or deaths accounted for 22.8%of all injuries. The shipbuilding workers are often uncomfortable working in a limited space with high and sharp workplaces. It means the seriousness of the risk factor of the shipbuilding work environment. Therefore, efforts to improve the working environment should be made continuously.

According to logistic regression analysis, if the skilled workers were injured, the possibility of becoming a death or a disability was 1.355 times higher than that of a novice. The more the work of experience, the older the age. Older experienced workers are less likely to suffer accidents, but if they do, they are more likely to become death or disability. Fraade-Blanar et al. [25] reported that older workers aged > 60 years had lower injury rates relative to younger and middle-aged workers. But, if injuries occurred, they were more severe and costly. Foreign workers were also 1.685 times more likely to be death or disability due to accidents than Koreans. On the other hand, regular workers showed a 1.362 times higher chance of becoming a death or a disability than non-regular employees. Among the types of accidents, it was found that an ‘explosion/burn’ accident is 7.158 times more likely to be a death or a disability than an accident caused by slip. Also, rupture injuries were more likely to be deaths or disabilities among the type of injuries, and facial injuries in the injured part of the body were more likely to deaths or disabilities than other parts of injuries.

5 Conclusion

This study presents an overview of actual nationwide compensation records that occurred in the shipbuilding industry. Of the total number of injured workers, 62.8%were accidents by novice less than one year of work experience. The proportion of novice was high in the age group over 60, or at small businesses with less than 15 employees. Also, the prevalence of accident types caused by novices was high in ‘fall,’ ‘cut/bruise,’ or ‘struck by.’ While the portion of source of accident caused by novices was high in ‘manhole,’ ‘ladder,’ or ‘scaffold.’

On the other hand, if the skilled workers were injured, they were more likely to be a death or a disability than a novice. If the face was injured, the possibility of being death or disability was higher than that of a leg, trunk, or neck/shoulder part of the body. Also, if the injury was caused by the rupture, it was found that the possibility of being death or disability was higher than other types of injury.

6 Limitation and implications of findings

This study has some limitations. First, the accidents reported in this study resulted in more than 4 days of absence from work. Hence, the accident rates reported in this study may show a lower frequency. Also, this study is a descriptive retrospective study of injury records. Thus, further research is required to investigate in-depth and qualitative relationships.

Nevertheless, it is meaningful that this study systematically derives the necessary information to prevent novice and death/disability accidents in the shipbuilding industry. The result of this study seems to be meaningful as the underlying data for accident prevention of the shipbuilding industry.

The results of this study suggest that safety education for novice workers should be systematically strengthened to focus on the elderly, female, foreign workers, temporary workers, and small businesses. Also, the safety education for novices, focusing on ‘manhole,’ ‘ladder,’ or ‘scaffold’ work, and ‘fall,’ ‘cut/bruise,’ or ‘struck by’ type of accidents, seems to be useful in accident prevention.

The results of this study suggest that attention for death/disability accidents should be paid to the improvement of the facilities and the wearing of the safety equipment in the work of skilled workers, explosion risk type of work, and facial risk type of work. There is a continuing need for efforts to reduce hazardous work conditions for these types of work and workers.

Footnotes

Acknowledgments

This research was financially supported by Hansung University.

Conflict of interest

None to report.

References

Statistics Korea, Korean standard industrial classification (KSIC). [Internet]. Daejeon: Statistics Korea; 2017. [cited 2019 Apr 24]. Available from: http://kssc.kostat.go.kr/ksscNew_web/ekssc/main/main.do&num;

OSHA. Shipyard employment eTool: shipbuilding. [Internet]. Washington, DC: Occupational Safety & Health Administration; 2019. [cited 2019 Apr 24]. Available from: https://www.osha.gov/SLTC/etools/shipyard/index.html

International Labour Office. Ship and boat building and repair: general profile. Encyclopaedia of occupational health and safety. 4th edition. [Internet]. Geneva: International Labour Office; 2011. [cited 2019 Apr 24]. Available from: https://www.iloencyclopaedia.org/contents/part-xv-26011/ship-and-boat-building-and-repair

You

, Jeong

, Lee

. Comparison of occupational accidents and illnesses in shipbuilding industry. J Ergon Soc Korea. 2019;38(5):385–92.

Pyo

, Jeong

. An implementation case of ergonomics program at a shipbuilding company. J Ergon Soc Korea. 2007;26(3):45–52. doi: 10.5143/JESK.2007.26.3.045

Barlas

. Shipyard fatalities in Turkey. Saf Sci. 2012;50, 1247–52.

Pyo

, Jeong

. Correlation analysis between safety and health indices in a shipbuilding industry. J Ergon Soc Korea. 2010;29(6):897–905.

Kim

, Ergonomic research trends in the shipbuilding industry: Present and the future. J Ergon Soc Korea. 2012;31(1):131–5.

, Jeong

. Risk factor analysis and ergonomics improvements in a shipbuilding industry, J Ergon Soc Korea. 2005;24(1):27–35.

10.

Jeong

, Kim

, Jeong

. Risk assessment in the shipbuilding industry: Present and the future. J Ergon Soc Korea. 2012;31(1):143–9.

11.

Lee

, The characteristics of industrial accidents in shipbuilding industry. J Ergon Soc Korea. 2012;31(1):137–42.

12.

Ministry of Employment and Labor, Industrial Accident Statistics 2016. [Internet]. Sejong: Ministry of Employment and Labor; 2017. [cited 2019 Apr 24]. Available from: http://www.moel.go.kr/info/publict/publictDataView.do?bbs_seq=20171200138

13.

Ministry of Employment and Labor, Industrial Accident Statistics 2017. [Internet]. Sejong: Ministry of Employment and Labor; 2018. [cited 2019 Apr 24]. Available from: http://www.moel.go.kr/info/publict/publictDataView.do?bbs_seq=20181200731

14.

Barlas

, Izci

. Individual and workplace factors related to fatal occupational accidents among shipyard workers in Turkey. Saf Sci. 2018;101:173–9.

15.

Jeong

, Kim

, A case study on the effect of ergonomics program in shipbuilding industry during the last ten years. J Ergon Soc Korea. 2012;31(1):191–6.

16.

Chen

, Chiou

. Comparison of static lifting capacity between experienced and novice Taiwanese female workers. Work. 2015;52(3):549–58.

17.

Thébault

, Gaudart

, Cloutier

, Volkoff

. Transmission of vocational skills between experienced and new hospital workers. Work. 2012;41(2):195–204.

18.

Warren

, Working culture: the agency and employment experiences of nonunionized workers in the surfboard industry. Environment and Planning A. 2014;46:2300–16.

19.

Korea workers’ compensation & welfare Service (COMWEL), COMWEL service guide 2017, [Internet]. Ulsan: COMWEL; 2017. [cited 2019 Apr 24]. Available from: https://www.kcomwel.or.kr/eng/comm/news.jsp

20.

Kim

, Jeong

. Occupational accidents and human errors in apartment custodians’ work. Work. 2018;60(4):587–95.

21.

Yang

, Jeong

. Comparison of accident characteristics between manual materials handling (MMH) and non-MMH works in the automobile parts manufacturing industry: Based on South Korea and the US. Work. 2019;62(2):197–203.

22.

Byun

, Park

, Jeong

. Effects of age and violations on occupational accidents among motorcyclists performing food delivery. Work. 2020;65(1):53–61.

23.

Hosmer

, Lemeshow

, Sturdivant

. Applied Logistic Regression. 3rd ed. Hoboken (NJ): JohnWiley & Sons, Inc. 2013.

24.

Lagac'e

, Nahon-Serfaty

, Laplante

. Canadian government’s framing of ageing at work and older workers: Echoing positive ageing models. Work. 2015;52(3):597604.

25.

Fraade-Blanar

, Sears

, Chan

KCG

, et al. Relating older workers’ injuries to the mismatch between physical ability and job demands. J Occup Environ Med. 2017;59(2):212–21.