Early retirement intentions of Korean wage earners: the influence of job demand-control-support latent profiles.

Background: The global aging of the population is a serious issue, and with working life expectancy increasing, it is crucial to consider measures to delay retirement. Since retirement intention is a key factor in determining the timing of retirement, understanding the psychological state of workers as a determinant of retirement intention is necessary. The purpose of this study is twofold: (1) to identify latent profile types based on job demands, control, and support among Korean wage earners using a person-centered approach; and (2) to examine the association between these latent profiles and early retirement intentions.

Methods: We analyzed data from 31,587 wage-earning participants aged 19 to 59 using the sixth Korean Working Conditions Survey (KWCS), conducted between 2020 and 2021. The sample included 57.06% men and 42.94% women. Latent Profile Analysis (LPA), a person-centered statistical method used to identify unobserved subgroups within a population, was employed to classify participants into five job characteristic profiles based on the job demand-control-support (JDCS) model. Job demands were measured across physical, quantitative, emotional, and social aspects; job control was assessed by items related to autonomy in task execution; and job support included perceived support from supervisors and coworkers. Early retirement intention, the outcome variable, was measured by asking participants the age until which they intended to work. Multivariate logistic regression analysis was conducted to examine the association between the identified job profiles and early retirement intentions, adjusting for relevant sociodemographic and occupational covariates.

Results: Five latent profile types were identified based on levels of job demands, control, and support using LPA. These profiles were labeled according to the Job Demand-Control-Support (JDCS) model and named as follows: Low Strain Collective (5.52%), Active Collective (27.99%), Passive Collective (28.92%), High Strain Collective (32.56%), and Low Strain Isolated (5.01%). The names reflect the distinct combinations of job demand, control, and support characteristics within each group. Multivariate logistic regression analysis showed that, compared to the Low Strain Collective, the Active Collective (OR = 1.65, 95% CI = 1.10-2.48), Passive Collective (OR = 1.72, 95% CI = 1.15-2.60), and High Strain Collective (OR = 1.66, 95% CI = 1.10-2.49) groups had significantly higher early retirement intentions. Additionally, gender, age group, education level, household income contribution, occupation type, employment type, and presenteeism were significantly associated with early retirement intentions.

Conclusion: Our findings suggest that to reduce early retirement intentions, employees should be given jobs that consider their personal and work characteristics, and they should have an appropriate level of job control. Moreover, creating a supportive atmosphere from supervisors and coworkers is essential.