This study examined the differences in the preference for long-term care (LTC) by age, period and cohort (A-P-C) in Japanese older adults through repeated cross-sectional surveys from 1998 – before the establishment of LTC insurance – to 2016, in a suburban city of metropolitan Tokyo. We analysed the direct effects of A-P-C on the preference for LTC, as well as the interaction effects of A-P-C on preference by gender, family structure and activities of daily living. Data were obtained at six time-points using repeated cross-sectional surveys for people aged 65 and older; surveys were conducted in 1998, 2002, 2004, 2010, 2013 and 2016. The preference for LTC was composed of three categories: informal care, community LTC services (CLTCS) and institutional LTC services (ILTCS). The cross-classified random-effect model was used to specify A-P-C effects. Informal care, CLITCS, ILTCS and other/no answer composed 35, 23, 33 and 9 per cent of preferences, respectively. In terms of the period effect, while there was an increase in levels of preference for CLTC between 1998 and 2010 as compared to informal care, the levels of preference were almost identical after 2010. In terms of the age effect, younger participants were more likely to prefer CLTCS and ILTCS over informal care. Moreover, the age influence was stronger in females and respondents who lived alone. We did not observe a cohort effect for preference. This study suggests that there are gaps by period and age between the preference for LTC services and the actual LTC use in Japanese older adults, and as a result, the use of actual LTC services cannot fully reflect the intentions and preference for LTC in them.