Effect of professional certification on employees’ return-to-work rate after occupational injuries in Korea: focusing on vulnerable groups

Background One effective way to improve return-to-work (RTW) performance may be to convince the employer that the worker has the necessary skills. The aim of this paper is to investigate the effect of having a professional certification among workers injured in occupational injuries on their return to work. Methods The Panel Study of Workers’ Compensation Insurance (PSWCI) targets workers who completed medical care in 2012 after an occupational injury. The study population (n = 2000) was stratified by gender, age, region, disability grade, and rehabilitation service use. A total of 1458 workers were finally selected for this study. The effect of having a certification on RTW status was calculated with an odds ratio and 95% confidence intervals using binomial and multinomial logistic regression analyses. In the binomial logistic regression analysis, the RTW group was made up as a combination of the return to original work and the reemployment groups. Results The ORs of RTW among those with a certification compared to those without certification were 1.38 (1.16–1.65) in Model 1, 1.25 (1.05–1.50) in Model 2, and 1.22 (1.01–1.47) in Model 3. Among female workers with a certification, the OR of RTW was 4.60 (2.68–7.91), that of return to original work was 3.21 (1.74–5.91), and that of reemployment was 5.85 (3.34–10.27). Among daily workers with a certification, the OR of RTW was 1.32 (1.03–1.69) and that of reemployment was 1.37 (1.07–1.76). Conclusion In conclusion, injured workers with a certification generally had a higher RTW rate. In particular, the RTW rate was higher among female workers and daily workers with a certification than among those without.


Introduction
The Korean Ministry of Employment and Labor reported in 2018 that the number of occupational injured workers who consequently required the minimum 4 days of leave was approximately 100,000 [1]. This figure represents about 0.54% of 19 million workers who were eligible for Workers' Compensation Insurance [1]. Even though this figure increased from 2017 to 2018, the overall trend of occupational injured workers has decreased [2]. However, regardless of such slight rises and falls, the total number of injured workers remains at a high level [1,2]. In addition, the economic loss due to occupational injuries is significant [3,4]. The direct compensation provided for occupational injuries was estimated at KRW 5 trillion (4.4 billion USD), and the estimated total economic loss, including indirect compensation, was KRW 25 trillion (21.8 billion USD) [1,3]. Furthermore, workforce loss due to occupational injuries and economic losses are closely related [4][5][6][7].
Korea has implemented the Industrial Accident Compensation Insurance Act to provide prompt and fair compensation and support for rehabilitation and social return of disaster workers in the wake of a disaster [4]. The Korean Workers' Compensation and Welfare Service (KCOMWEL) provides disability benefits to injured workers to compensate for the loss of income due to an accident [8,9]. For individuals who are injured and not working, receiving immediate financial support and rehabilitation is critical. However, these forms of primary support are not enough to ensure employees maintain their previous work performance levels after the accident. The accident causes injured workers not only to lose the ability to work but also to experience loss of self-esteem and self-efficacy and depression, which hinders their return-to-work (RTW) rate [4,[10][11][12][13][14][15].
Several studies have identified factors affecting the RTW rate of injured workers [16][17][18][19]. These factors can be divided into five categories: personal, socioeconomic, occupational, degree of injury, and employer interest level in workers' reemployment [4,16,19]. RTW is known to be associated with personal factors such as being male, being younger, being married, having a higher education level, and having a higher household income [18] and occupational factors such as larger-sized workplaces, longer duration of employment, regular workers, and other factors such as lower disability ratings after occupational injuries, shorter length of hospital stays and recovery duration, hospitals with higher quality scores, and greater interest of doctors in charge and employers in injured workers' RTW [20,21]. However, the above factors are outside the individual's control and are hard to change even if the workers have strong will to RTW. Thus, it is necessary to study the factors within injured workers' control that can positively affect RTW to assist in their RTW preparation.
One effective way to improve RTW performance is to convince the employer that the worker has the necessary skills. Professional certification in the field is often used as an official permit for performing a job [22,23], and it can provide employers an objective evaluation of employees' skills [22,24]. Therefore, it has been recognized as a useful resource to help someone get a job [25][26][27]. It is possible that as an indicator of skills, professional certification could assist injured workers who want to RTW.
Most studies on occupational injuries focused on RTW [9,18,19]. However, few have examined the relationship between workers' RTW and professional certification. Therefore, this study investigates the effect of professional certification on injured workers' RTW after occupational injuries.

Study design and participants
In this study, we used data from the first cohort (firstfifth waves) survey of the Panel Study of Workers' Compensation Insurance (PSWCI) conducted by the Korea Workers' Compensation and Welfare Service (KCOM-WEL). The PSWCI provides useful data on all policy processes related to Industrial Accident Compensation Insurance Act, including those involving injured workers returning to work. The PSWCI targets workers who completed medical care in 2012 after an occupational injury (n = 82,493); the study population (n = 2000) was stratified by gender, age, region, disability grade, and rehabilitation service use.
Since the first survey in 2013, the PSWCI has been conducted annually. The fifth wave was completed in 2017. The study methods comprised one-to-one interviews conducted by professional interviewers who visited the participants; participants were asked to respond to the questionnaire in person [4,[28][29][30][31][32].
Of the total 2000 participants in the first cohort survey, 1616 participated. Participants who had participated in the first-fifth waves were included in the present study. One hundred and two workers who had not continuously participated in the first-fifth waves were excluded. In addition, five workers who were selfemployed or employers at the time of the occupational injury and 51 injured workers whose explanatory variables were not observed were also excluded. Therefore, a total of 1458 workers were finally selected for this study (Fig. 1).

Occupational-related characteristics
Occupational factors included industry sector, status of workers, number of employees, and duration of employment in the workplace at the time of occupational injury. Industry sector was categorized according to the Korean Standard Industry Classification (KSIC) (based on the International Standard Industry Classification, ISIC) into "manufacturing" and "construction"-which accounted for more than half of occupational injuries and diseases (manufacturing: 26.8%, construction: 27.1% in 2018 [1])-and "services (Information and communications; professional, scientific and technical activities, etc.)" and "other (Agriculture, forestry and fishing; Mining and quarrying, etc.)." For worker status, full-time workers were classified as "regular workers," whereas temporary and daily workers were classified as "daily workers." The "number of employees in the workplace" was divided into four groups: < 5, 5-9, 10-29, and ≥ 30. The "duration of employment at the workplace" was divided into three groups: < 1 year, 1-3 years, and ≥ 3 years.

Injury-related characteristics
Data on occupational injury type and disability rating were obtained from the KCOMWEL's administrative database. The type of occupational injury was classified into injuries and diseases. If workers suffered any mental or physical disability from occupational-related injuries or diseases and met the requirements for industrial accident compensation, they were assigned a disability rating from Grades 1 to 14 under the Korean Industrial Accident Compensation Insurance Act. A grade close to Grade 1 indicates a more serious disability [4,28]. Occupational injuries were classified into three categories: 1-7 (severe), 8-14 (moderate), and "none."

Qualifications-related characteristics
We investigated whether participants held any national certification (professional engineer, engineer, master craftsman, industrial engineer, craftsman, or other national certification), private degrees, or international (foreign) certificates (excluding general driving license). In the first wave of the panel study, participants with one or more certifications were coded as "yes," and those with no certification were coded as "no."

Main outcome variables
The type of economic activity included return to original work, reemployment, self-employment, unpaid family work, unemployed, or economically inactive. Those who returned to their original work were workers who were injured at the workplace, but resumed work in the same position they held at the time of injury. Workers who found paid employment in a workplace other than where they experienced the injury were referred to as reemployed workers. Self-employed workers included private business owners or freelancers, and unpaid family workers referred to those who helped their family or relatives in the workplace for an average of > 18 h per week (more than 3-4 h per day). Unemployed persons referred to those who reported searching for a job more than once during the past 4 weeks and were able to work if they found a suitable job during the last 1 week; those who did not were classified as an "economically inactive population [20]." In this study, the type of economic activity was classified as "return to original work," reemployed and selfemployed persons as "reemployed," and unpaid family workers, the unemployed, and the economically inactive population as "non-return-to-work" (non-RTW).

Statistical analyses
The characteristics of the participants regarding work status, and of those who returned to work with a certification, were analyzed using a chi-square test. The effect of having a certification on RTW status was calculated with an odds ratio (ORs) and 95% confidence intervals using binomial and multinomial logistic regression analyses. In the binomial logistic regression analysis, the RTW group was combined with the "return to original work" group and the "reemployed" group. In the multinomial logistic regression analysis, the non-RTW group was compared with the "return to original work" group and the "reemployed" group. All analyses were performed using the software SAS statistical package version 9.4 (SAS Institute, Cary, NC, USA).

Results
The general participant characteristics according to RTW status are shown in Table 1. The return-to-original-work rate after occupational injuries was 38.1%, the reemployment rate was 51.2%, and the unemployment rate was 10.6% (P < 0.0001). The reemployment rates among male and female workers (51.6% and 49.4%, respectively) were higher than those for other RTW status (P = 0.0002). Regarding worker status, regular workers were found to have a higher rate of return to original work, whereas daily workers were found to have a higher reemployment rate (P < 0.0001). Those with a certification had a 42.7% return-to-original-work rate and a 50.3% reemployment rate (P = 0.0001).
The general characteristics of RTW workers were based on whether they held a certification ( Table 2). The RTW rate among those with a certification was 93.0%, whereas the RTW rate among those without a certification was 86.9% (P < 0.0001). Overall, the RTW rate was higher among those with a certification except for those aged less than 30, with a marital status of "other," educational level of "College graduate or above," occupational injury type of "Disease," duration of employment of 1-3 years, and disability rating of Grades 1-7. Tables 3 and 4 show the ORs of RTW according to whether the participant had a certification. Model 1 was adjusted for gender and age. In Model 2, adjustments were made for the variables in Model 1 plus marital status, educational level, type of industry, worker status, number of employees, and duration of employment. Model 3 was adjusted for the variables in Model 2 plus disability rating and type of occupational injury. The ORs of RTW among those with a certification compared to those without certification were 1. 38 (Table 4). Table 5 shows the ORs of RTW based on having a certification, which was stratified by age and status of workers. Among female workers with a certification, the OR of RTW was 4.60 (2.68-7.91), that of return to original work was 3.21 (1.74-5.91), and that of reemployment was 5.85 (3.34-10.27). Among daily workers with

Discussion
This study investigated the effects of holding a certification on RTW among occupationally injured workers after completion of medical care. The ORs of RTW (4.60, 2.68-7.91), return to original work (3.21, 1.74-5.91), and reemployment (5.85, 3.34-10.27) among female workers with a certification were higher than in those with no certification. In addition, the ORs of RTW (1.32, 1.03-1.69) and reemployment (1.37, 1.07-1.76) among daily workers with a certification were higher than those with no certification (Table 5). Thus, the effect of having a certification on RTW is greater among female workers and daily workers who were relatively more vulnerable to a lower RTW rate than male workers and full-time workers, who are known to have a high RTW rate. This finding is consistent to the results of a previous study, which reported that having a certification had a positive effect (Coef. 1.468) on employment among vulnerable groups [27]. We therefore believe that having a certification is crucial for RTW in injured workers, especially those belonging to vulnerable groups.
Among sociodemographic factors, a higher educational level was found to be associated with a higher return-tooriginal-work rate. In terms of occupational factors, the return-to-original-work rate was higher among those who were in manufacturing industries, those who are regular workers, those whose workplace had more than 30 employees at the time of the occupational injury, and those who had worked for more than 3 years at the workplace where the occupational injury occurred. In terms of injury-related factors, non-disability or a lower disability grade was associated with a higher return-tooriginal-work rate (Table 1). This is consistent with previous findings in that the rate of return to original work was higher among students with high school graduation or more, manufacturing workers, regular and full-time workers, workers of workplaces with at least 50 employees at the time of occupational injury, workers whose duration of employment at the workplace at the time of occupational injury was 1 year or longer, and workers with no disability or with grade 10-14 disability [33][34][35][36][37]. In particular, the return-to-original-work rate among construction workers after an occupational injury was lower than among manufacturing and service workers. This was likely due to a higher percentage of daily workers in the construction industry than in the manufacturing and service industries [4,38]. In addition, most construction workers were older and thus more likely to have a disability after a workplace accident [4].
With regard to the association between RTW and holding a certification, the RTW rate among workers with a certification was higher than among those without a certification ( Table 2). The ORs of RTW among those with a certification were higher in Model 1 (1.38, 1.16-1.65), Model 2 (1.25, 1.05-1.50), and Model 3 (1.22, 1.01-1.47), indicating that the ORs of RTW in all models were higher among those with a certification, compared to those without one (Table 3). In addition, the OR of return to original work was 1.47 (1.22-1.77) in Model 1 (Table 4). This is consistent with the finding that RTW rate was higher among workers with a certification than among those without a certification [22,[24][25][26]. Therefore, it is clear that having a certification can affect RTW among injured workers. Although the KCOMWEL provides injured workers with customized rehabilitation support services across stages of medical care to facilitate their RTW and help the community at large [1][2][3], no support is offered for acquiring  certification. We recommend that, as with vocational training support and employment support for injured workers, additional certification support should also be provided.

Limitation and strength of the study
This study has several strengths. First, it is significant in that the results provide comprehensive data on various situations and needs of injured workers because they deal extensively with RTW and socioeconomic characteristics of injured workers. In addition, data from the only panel study in South Korea that includes occupationally injured workers were used. Second, the study examined the relationship between having a professional qualification and RTW and, for the first time, attempted to investigate the relationship among injured Korean workers.
The limitations of the study were as follows. In the PSWCI, eight types of certifications (national, private, and international) were surveyed. However, as the number of persons with each type of certification was low, separate analysis regarding the effects of each type was not feasible. Therefore, analysis was performed based only on whether someone had a certification. However, RTW among injured workers may have been affected by certification type. The data used in this study were classified according to the "Yes" or "No" of a certificate in the first wave of the PSWCI. The PSWCI did not investigate the year of obtaining a certificate, so they may have obtained a certificate after an occupational injury. The time of obtaining the certificate could affect the return-to-work date. In addition, as the analysis was based on PSWCI data obtained through household interviews during a specific period, recall bias is possible [4,28,30,32].

Conclusions
In conclusion, injured workers with a certification generally had a higher RTW rate. In particular, the RTW rate was higher among female workers and daily workers with a certification than among those without. This implies that having a certification may play a more important role in RTW among female workers and daily workers, who were relatively vulnerable to lower RTW rates. Therefore, strategies are needed to assist injured workers, especially those belonging to vulnerable groups, in obtaining certifications that facilitate their RTW.