Food insecurity affects many dimensions of well-being. It may lead to under nutrition. Although geriatric subjects from food insecure households are more likely to have inadequate energy, household food security does not necessarily guarantee for energy adequacy at individual level. This community based cross-sectional study was conducted on 616 geriatric subjects to find out predictors of food insecurity at household and individual level and examine energy adequacy at family versus individual level. Subjects were selected by multistage sampling procedure from urban Varanasi, India. A predesigned, pretested proforma was used for collecting socioeconomic information. Food insecurity was assessed using Household Food Insecurity Access Scale; 24 hours recall method was used for energy intakes. For analysis of data SPSS version 22.0 was used. χ2 and logistic regression were used for inferential purpose. Food insecurity at household was 68.2%.The significant association of marital status, religion, type and size of family and literacy status with household food insecurity in univariate analysis got eliminated after applying logistic regression analysis; Adjusted Odds Ratios were higher for SC/ST Caste (2.13; 95% CI: 1-4.5) and upper lower plus lower socio economic status(24.13; 95% CI: 13.87-44.61). On energy intake considerations, of 53.9% subjects with household food insecurity, 80.7% had individual food insecurity whereas of 46.1% subjects with household food security 63.7% subjects had individual food insecurity. Caste and socioeconomic class predicted food insecurity at household level. Energy adequacy at family level did not ensure energy adequacy in geriatric subjects.