Several studies indicate that older age is a risk factor for probable post-traumatic stress disorder (PTSD). However, evidence on the prevalence and risk factors for elderly disaster survivors is limited. Multi-stage stratified sampling was applied in this cross-sectional study. The Revised Version of the Impact of Event Scale (IES-R) was used to evaluate symptoms of PTSD. Chi-squared test was used for univariable analysis of prevalence of probable PTSD by possible determinants. Multivariable logistic regression analysis was utilized to further explore risk factors for PTSD. 427 elderly survivors age 60 or older participated. The prevalence of probable PTSD was 40.5, 36.3, and 21.5% according to a cut off score of the IES-R of 33, 35, and 2 points on average across items, respectively. In multivariable logistic regression, elderly survivors with a higher number of diagnosed chronic illnesses were more likely to be screened positive for PTSD and those with a greater number of family members living in the same home were less likely to be classified as having probable PTSD for all cut off scores. Elderly survivors with improved economic status and those with primary school or lower education were more often estimated to have probable PTSD with a cut off score of 35 and 2 points on average. Elderly survivors in a high impact area following a major earthquake are at high risk for developing PTSD. Those who live alone and report pre-existing chronic illness are particularly vulnerable.