The Ministry of Health, Labour and Welfare has conducted a comprehensive survey of living conditions in Japan every year since 1986 to help evaluate living conditions, welfare, health, and income; to assist with planning, management and policy implementation . Conducted every three years, this survey covers approximately 289,363 households nation-wide in randomly selected areas. Participants are randomly selected from 5,150 municipalities of the National Census. In designated municipalities, instructors at public health centers in prefectures conduct training for persons who are in charge of the survey and conduct interviews with selected households. To conduct the present study, we requested data from the 2010 database according to procedures governing the use of official MHLW statistics. The file included de-identified data on all family members from each of the 228,664 households who had agreed to participate in the survey during 2010. We analyzed data from all men aged 20–59 years who were at the time, employed and the head of a household whose earnings maintained family.
Questions in the MHLW survey included basic demographic information, such as age, sex, occupation, and employment status. There were 13 categories of occupation: 1) management, 2) professional and technical work including teachers, health care workers and researchers (professional), 3) clerk, 4) sales, 5) service, 6) security, 7) agriculture and fishery (agriculture), 8) manufacturing, 9) transportation or machine operator (transport), 10) construction, 11) cleaning, packing operators (cleaning), 12) others, and 13) unknown. Employment status consisted of self-employed, employers of companies, regular workers, non-regular workers, and others .
The first question regarding the problems on present personal problems causing annoyance and stress in life was: “Do you have any problems that make you annoyed or feel stress in daily life at present (Yes/No)?” The people who answered “Yes” were asked to select all possible causes of annoyance or stress from 21 categories, and then to select the most concerning present personal problems: 1) personal relationships with family, 2) personal relationships with anyone except family, 3) love or sex, 4) marriage, 5) divorce, 6) bullying or harassment, 7) Concerned about one’s purpose in life, 8) no time for myself, 9) financial difficulties, 10) own diseases, 11) disease and care of a family member, 12) pregnancy of their wife, 13) child-rearing, 14) household chores, 15) Education for himself, 16) education of his children, 17) own tasks or duties of work, 18) family’s working conditions, 19) housing and environmental conditions, 20) others, and 21) do not know. Some respondents did not choose any items from this list even though they had answered “yes” for the first question asking if they had any problems that made them annoyed or stressed in life.
The questionnaire also included the Japanese version of the Kessler 6 (K6) which is comparable with the center for Epidemiologic Studies Depression Scale (CES-D) [14,15]. The cut-off of the K6 for determining psychological distress was over 5, based on a previously validity study of a Japanese version of the K6 with 100% sensitivity, 68.7% of specificity based on the Youden index which provided the same optimal cut-off point for the scale .
We first conducted chi-square analyses to determine the associations between psychological distress and each variable of the most concerning present personal problem. Logistic regression was used to examine potential associations of the most concerning present personal problems with psychological distress, adjusting for occupational variables (occupation and employment status) and age, using problems from own tasks or duties of work as the reference [16,17]. Logistic regression calculated the odds ratios (OR) and confidence intervals (CI). All analyses were performed using IBM SPSS Statistics 20, with the level of statistical significance set at p < 0.05. All ORs were adjusted using Zhang’s correction formula for common outcomes, given that the prevalence of psychological distress was relatively high .
This study involved a retrospective analysis of data that had already been obtained during a national survey. As we did not use any personally identifiable information, based on regulations existing in Japan, ethical approval was not required.