This cross-sectional study involved police officers from 15 regional police departments (Wonju, Sokcho, Hwacheon, Hoengsung, Yanggu, Chuncheon, Pyeongchang, Jungsun, Gangneung, Yeongwol, Donghae, Inje, Goseong, Hongcheon, and Taebaek) in Gangwon from June 4 to December 20, 2015. The study was initially conducted on 272 police officers among 3461 police officers in Gangwon, by convenience sampling, study subjects who have not experienced traumatic events for the past 6 months and carried out rotation duties between patrol job and office work. They were all aged over 49 years, have been performing rotation duties between patrol and desk job for more than 20 years, but all of them were engaged in patrol jobs as the present task. All of them visited the Department of Occupation and Environmental Medicine, Wonju Christian Hospital, for routine health check-ups, special screening for work-shifts, PTSD symptoms, and depression. All participants completed a self-reported questionnaire for evaluating job-related stress, depression, self-resilience, and PTSD symptoms. Among 272 participants who completed the survey, answers of 39 were incorrect, and 13 were previously diagnosed with mental disorders or had familiar history of mental disorders; we excluded such participants. Among the 220 remaining police officers, 112 had experienced traumatic events and were finally enrolled as study subjects. Written informed consent was obtained from participants prior to their involvement in the study.
Basic demographical information was collected by self-reported questionnaires that included the following: age range, 47 to 60 years (the data were processed as continuous data), education status (middle school or below, high school, or college or above), marital status (married, divorced, or widowed), smoking status (non-smoker, ex-smoker, or current smoker), and frequency of drinking alcohol (none, ≤ one drink per week, or ≥ two drinks per week).
The occupational characteristics surveyed included the following: service area (urban or rural); years of patrol service (<10 year, 10–19 years, ≥ 20 years), and job stress (measured using the Korean Occupational Stress Scale-Short Form [KOSS-SF]).
Twenty-four questions on the 4-point Likert scale from 1 (“not at all”) to 4 (“very much”) were included in the KOSS-SF, which was validated using factor analysis and standardized validation process by the National Study for Development and Standardization of Occupational Stress and included the following seven subscales: job demand (four items), insufficient job control (four items), inadequate social support (three items), job insecurity (two items), organizational system (four items), lack of reward (three items), and occupational climate (four items). The sum of the conversion scores for each of the seven sectors was divided by seven to obtain the total occupational stress score. Based on the reference values of occupational stress for men (Short Form) , the total study groups were categorized by quartile to observe the score distribution of occupational stress, and the data were processed as continuous data to conduct t-test and logistic analysis.
To evaluate the subjects’ psychosocial levels, the Korean Center for Epidemiologic Studies Depression Scale (for depression) (K-CES-D) and Connor-Davidson Resilience Scale-Korea (CD-RI-K) (for self-resilience) were used.
Korean center for epidemiologic studies depression scale (K-CES-D)
To measure depression, which was expected to be associated with PTSD symptoms, the adapted Korean version of the Center for Epidemiologic Studies Depression Scale (CES-D) was used. It consisted of 20 questions based on symptoms of depression that were experienced during the week; the answers ranged on a scale from 0 to 3 where 0, 1, 2, and 3 denoted very rarely, rarely, sometimes, and most of the time, respectively. Questions 4, 8, 12, and 16, where positive meanings were contained, were reversely calculated. Total score below 16, 16–20, 21–24, and over 25 indicated normal, mild, severe, and very severe, respectively. The reliability and validity of this questionnaire are well established .
To measure self-resilience, a Korean-version of the self-resilience scale (CD-RISC-K), which is an adapted version of CD-RISC  developed by Block and Klomen, was used. The scale included 25 questions in five areas regarding toughness, durability, optimism, support, and spirituality as subareas with a scale of 0 to 4, where 0, 1, 2, 3, and 4 represented absolutely not, not, usually, yes, and absolutely yes, respectively. The range of scores was 0 to 100. A higher score indicated higher resilience. In this study, the total scores were analyzed as continuous variables. In addition, based on the average score for self-resilience (64.5) of 552 firefighters in Korea in a previous study , two groups were formed for analysis: low resilience and high resilience groups. Several studies have established this questionnaire as a reliable and valid instrument for measuring symptoms of depression in the Korean population .
Critical incident exposure
Through a self-survey, we investigated the extent of exposure to traumatic events such as a disaster, an accident, physical, or sexual abuse (“experienced,” “witnessed,” “learned of,” “not sure,” and “doesn't apply”) while working. Those who experienced or witnessed a traumatic event while working were included in the final analysis.
Current PTSD symptoms
PTSD symptoms were measured using the Impact of Event Scale-Revised-Korean (IES-R-K).
The IES-R-K is a 22-item self-reporting method that assessed subjective distress caused by traumatic events. Items corresponded directly to 14 of the 17 Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) symptoms of PTSD. Respondents were asked to identify a specific stressful life event, and then, indicate how much they were distressed or bothered during the past 7 days by each “difficulty” listed. Items were rated on a 5-point scale ranging from 0 (“not at all”) to 4 (“extremely”). The IES-R yields a total score (ranging from 0 to 88), where IES-R score ≤24 = normal, 25–39 = mild/moderate symptoms of PTSD, 40–59 = severe symptoms of PTSD, and ≥60 = very severe symptoms of PTSD. In this study, anyone with 25 points or more based on IES-R-K score was defined as having PTSD symptoms.
To analyze the results, the chi-squared test, two-sample t-test, and logistic regression were used. We conducted a descriptive analysis on the general and occupational characteristics of the subjects surveyed. The crude odds ratios (OR) between the expected protective factors, that is, self-resilience and occurrence of PTSD symptoms and 95 % confidence intervals (95 % CI) were calculated using a simple logistic regression analysis. To identify the factors that might influence PTSD symptoms, factors were corrected for and were applied as independent variables, and PTSD symptoms were applied as the dependent variable in the logistic regression analysis. Model 1 was based on crude OR; model 2 included the socio-demographic characteristics as the independent variables and model 3 included the socio-demographic characteristics and occupational factors as the independent variables. Model 4 included the socio-demographic characteristics, occupational factors, and depression as the independent variables. OR and 95 % CI were calculated. All data were analyzed using SPSS version 22.0.