To the Editor: We read with interest the article published by Somadder et al.1 The authors document a correlation between depressive symptoms and self-reported numbers of falls in older subjects attending a day hospital in the United Kingdom. They reported that there were no significant differences in age, comorbidities, or performance on activities of daily living (ADLs) between fallers and infrequent fallers in their small population. We reexamined this important issue in community-dwelling elderly people in Japan and found findings different from those of Somadder et al. The study population consisted of 1,261 people aged 65 and older (men 529, women 732, mean age 75.4 ± 7.2) living in T town, Kochi Prefecture, Japan. Fallers were screened using self-reported questionnaires, along with additional tests of ADLs and subjective quality of life (QOL) for community-dwelling older people in 2006. The question “Do you have any history of a fall within the past year?” was used for detecting fallers. Subjects who answered yes to the question were considered to be fallers. For the assessment of basic ADLs, the scores for seven items (walking, ascending and descending stairs, feeding, dressing, using the toilet, bathing, and grooming) were summed using a rating scale from 0 (completely dependent) to 3 (completely independent) to obtain a basic ADL score (0–21). For advanced ADLs, the Tokyo Metropolitan Institute of Gerontology index of competence rating scale of 0 to 13 was used.2 This scale includes instrumental self-maintenance (0–5), intellectual activity (0–4), and social role (0–4). Five indicators of QOL (sense of subjective health, relationship with family, relationship with friends, financial satisfaction, and subjective happiness) were rated on a 100-mm visual analogue scale (worst QOL on the left end of the scale, best to the right).3,4 The 15-item Geriatric Depression Scale (GDS-15)5 was used for the assessment of depression; a score of 10 or more was considered to indicate depression. A fall risk index6,7 with a score ranging from 0 (low risk of fall) to 21 (high risk of fall) was added to those and used for the assessment of risk of falls. Statview version 5.0 (SAS Institute, Inc., Cary, NC) was used for calculating chi-square tests for categorical variables, unpaired t-test for continuous variables, and Spearman correlation (rs) between number of falls and GDS-15 and between fall risk index and GDS-15. The proportion of fallers was 31.6% in this population. Fallers were significantly older (76.9 vs 74.7) and had significantly lower scores for each item of the ADLs and QOLs than nonfallers, even after the adjustment for age (Table 1). The proportion of subjects with depression was significantly higher in fallers (26.8% vs 11.6%, P<.001). Although only 59.6% of the fallers answered the numbers of falls, there was weak but significant correlation between number of falls and GDS-15 scores in those who had fallen (rs=0.17, P=.002). The mean fall risk index score was significantly higher in fallers than nonfallers, and there was significant correlation between fall risk index and GDS-15 (rs=0.53, P<.001) in fallers. We confirmed the higher prevalence of depression in fallers than nonfallers, and there was a significant correlation between the number of falls and GDS-15, as Somadder et al. reported. However, unlike with the findings of Somadder et al., community-dwelling elderly fallers in Japan were significantly older and had lower quantitative ADL and QOL scores, as well as higher GDS-15 scores than nonfallers, even after adjustment for age. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that none of the authors have any financial or any other kind of personal conflicts with this article. Financial support for this study was obtained in part with funds from working group of fall prevention, Japanese Ministry of Health, Labor and welfare. Taizo Wada was supported by Global COE (Centers of Excellence) Program, In search of Sustainable Humanosphere in Asia and Africa (E-04), Japan Society for the Promotion of Science (JSPS). Author Contributions: Taizo Wada: data analysis and writing the manuscript; Masayuki Ishine, Yasuko Ishimoto, Mayumi Hirosaki, Yumi Kimura, and Yoriko Kasahara: data collection; Kiyohito Okumiya, Masanori Nishinaga, Kuniaki Otsuka, and Kozo Matsubayashi: study concept. Sponsor's Role: The sponsors did not participate in the design or data analysis of any aspect of the study or manuscript preparation.