A pilot study assessed the clinically determined and self-reported oral health status of 50 randomly selected homebound patients served by Boston's Home Medical Service. The sample was largely female, low-income, and edentulous. The median age of the patients was 81 years (range, 64-101). While 76% deemed themselves to be in good to excellent oral health, 80% of the patients had not seen a dentist within the last two years, and 80% were found to be in need of routine dental care. To assess whether the Geriatric Oral Health Assessment Index (Atchison and Dolan, 1990) could be used by non-dental health professionals to determine the need for requesting dental consultation, the study physician repeated the administration of the GOHAI for 23 of the 50 subjects within eight weeks of the initial examination. For the 23 subjects having both dentist- and physician-administered GOHAI scores, the intraclass correlation coefficient was r = 0.61 (p = 0.002), indicating good agreement between the dentist's and physician's administrations of the GOHAI. However, given the high prevalence of need for care, the GOHAI appears to be of less value than an examination for identifying persons who need dental care in this population. Future research is needed to examine the GOHAI's sensitivity and specificity in populations with low to moderate prevalence of treatment need.