Three tertiary-care hospitals, National Health Insurance Service Ilsan Hospital (817 beds), Ilsan Paik Hospital (800 beds), and National Cancer Center (605 beds), in Goyang City of Gyeonggi Province, which has a population of 1,005,102 persons, participated in the study. All HCWs who worked in those hospitals were eligible for the study. The vaccination records on 2017–2018 influenza vaccine, age, sex, and occupational characteristics of HCWs during the vaccination campaign in each hospital, were collected retrospectively. HCWs who were 65 years or older were excluded from the study because they were also eligible for the national immunization program of seasonal influenza vaccination. Volunteers were excluded from the study because of lack of personal-identifiable data and service-related information. The study was approved by the Institutional Review Boards of the National Cancer Center (IRB No. NCC-2018-0119) and of the other participating hospitals. Requirements for informed consent were waived. All participant-identifiable data were deleted or modified prior to data analysis.

The occupational groups were divided into medical and non-medical groups. The medical group is defined as the occupational group that specializes in medical practice, and non-medical group is defined as the occupational group that conducts medical assistance or general service. The medical group was further divided into physicians, nurses, and health technicians. In the case of physicians, they were subdivided into medicine/pediatrics, surgery, intern doctor, and others subgroups. Others group included the specialist doctors of emergent department, radiology, pathology, laboratory medicine, dermatology and ophthalmology. Nurses were also subdivided into nurses working in in-patient wards, out-patient clinics, and other subgroups, depending on their workplace. Healthcare technicians included emergency paramedical personnel, laboratory staffs, nursing assistants, pathology technicians, radiology technicians, and rehabilitation therapists. For non-medical groups, the HCWs were divided according to the degree of frequent contact with patients. Group A (frequent contact with patients) included security agents, patient transfer agents, janitors, and clinical trial-related researchers; whereas laboratory researchers, general office workers, medical records staffs, nutritionists, facility management staffs, and bioengineering staffs were included in group B (less-frequent contact with patients).

Prior to the 2017–2018 influenza season, influenza vaccinations were provided to all HCWs at each hospital. To increase the accessibility of vaccination, all hospitals have provided free influenza vaccinations through in-service carts to hard-to-reach staff, and medical HCWs in 2 out of the 3 hospitals were allowed self-vaccination in general wards. Each hospital encouraged the seasonal influenza vaccination through a campaign, educated, and promoted the need for vaccination; each hospital headquarters received reports on the HCW vaccination rates. In the case of on-site vaccination or mobile cart services, vaccines were kept in a cold box/freezer set with a thermometer (2°C and 8°C) until vaccination to maintain a ‘cold chain’. Table 1 summarizes the details of the campaign program of each hospital.

Categorical variables were summarized with frequencies (percentages). To evaluate the association between categorical variables and vaccination rate, χ² test was used for univariate analysis. A difference of at least 5% among the groups was considered to be different, and estimates with smaller differences were considered similar.6) P-value <0.05 was considered as statistical significance. Statistical analysis was executed using SPSS version 19 (IBM Corp., New York, NY, USA) and GraphPad Prism version 7.03 (GraphPad Software, La Jolla, CA, USA).

Among physicians, vaccination rate of the medicine/pediatrics subgroup was the highest with 88%, followed by others subgroup with 84%, surgery subgroup with 80%, and intern doctor subgroup with 77%. The vaccination rate of the medicine/pediatrics and others subgroups was 86%, which was significantly higher than that of the surgery and intern doctor subgroup (79%) (P=0.013, Fig. 2A). No difference in the vaccination rate among the age groups within the physician group was noted (P=0.94). Female physicians seemed to have a higher influenza vaccination rate than male physicians, but there was no significant difference (87% and 82%, P=0.063). Among nurses, the vaccination rate of the out-patient clinic and in-patient ward subgroups, who frequently contact and care for patients, was significantly higher than that of the other nurse subgroups (93% and 87%, P<0.001, Fig. 2B). In the subgroup analysis, the vaccination rate of in-patient ward subgroups was 92% and that of out-patient clinic subgroup was 96%, which was statistically significant (P=0.003). Vaccination rate was highest in nurses who were at least 50 years (99%), followed by 40–49 years (94%), 20–29 years (93%), and 30–39 years (87%) (P<0.001). Female nurses seemed to have a higher influenza vaccination rate than male nurses, but there was no significant difference (92% and 88%, P=0.288). Among healthcare technicians, no difference in the vaccination rate according to age or sex was noted (data not shown).

For non-medical HCWs, influenza vaccination rates were different, depending on the frequency of contact with patients. In the case of group A non-medical HCWs, who had frequent contact with patients, the vaccination rate (90%) was significantly higher than that (73%) of group B non-medical HCWs, who had less frequent contact with patients (P<0.001, Fig. 2C).