that, if the traditional threshold of aging were to be brought forward by, let’s say, 5 to 10 years, the challenges that are currently attributed to aging societies would disappear. Al- though such a proposal may today seem controver- sialFmostly for economic reasonsFwithin the next few decades, the numerous healthy and vital 65-year-olds who are not yet willing to retire may cause this solution to be- come inevitable. Krzysztof Ksiazek, PhD Department of Pathophysiology Poznan University of Medical Sciences Poznan ˜ , Poland ACKNOWLEDGMENTS I thank Janusz Witowski, MD, PhD, for helpful criticism and discussion. Conflict of Interest: No financial support received. No conflict of interest. Sponsor’s Role: There was no sponsor for this work. REFERENCES 1. Sehl ME, Yates FEJ. Kinetics of human aging: I. Rates of senescence between ages 30 and 70 years in healthy people. J Gerontol A Biol Sci Med Sci 2001;56A:B198–B208. 2. Davies AM. Epidemiology and the challenge of ageing. Int J Epidemol 1985;14:9–21. 3. Oeppen J, Vaupel JW. Demography. Broken limits to life expectancy. Science 2002;296:1029–1031. 4. Coles LS. Demography of human supercentenarians. J Gerontol A Biol Sci Med. Sci 2004;59:B579–B586. 5. Westendorp RG. What is healthy aging in the 21st century? Am J Clin Nutr 2006;83suppl:404S–409S. FAMILY SUPPORT OF ELDERLY PATIENTS HOSPITALIZED IN A PUBLIC TERTIARY HOSPITAL IN GREECE: A PROSPECTIVE STUDY To the Editor: Elderly people have needs that require other people’s help. If a companion does not exist, their quality of life becomes compromised, because their well-being de- pends strongly on family support, and neglect and loneli- ness may lead to depression. 1–4 Mediterranean families have strong bonds, and young- er family members give plenty of support to their elderly relatives. 5–8 Nevertheless, little information exists regard- ing changes of attitudes in these societies during the last decades, when traditions are vanishing. More specifi- cally, no data and predictors for support exist in Greek society. The present study evaluated the support provided by family and detected factors predisposing to neglect of el- derly patients hospitalized at the University Hospital of Heraklion, Crete, Greece. A pilot cross-sectional study was performed for a 6- month period including all elderly patients (aged  70) hospitalized in the Department of Medicine, after approval by the ethics board. Data were retrieved from medical re- cords; escorts, if any; and nursing staff and included age, reason for hospitalization, outcome, underlying disease and prognosis, previous hospitalization, place of living before and after present hospitalization, marital status, number of children, profession of children, socioeconomic status of the family, patient’s profession, source of income, patient’s ed- ucation, number of family members or companions visiting, time they spent with the patient, and care offered by rel- atives or companions. Data, converted into variables, were analyzed as risk factors possibly predisposing to absence of support. Patients were considered supported if at least one fam- ily member paid daily visits, and the remaining patients were considered neglected. The chi-square test was used to assess the association between different factors and support. Factors significantly associated with support, in univariate analyses, were en- tered in a multivariate backward, stepwise, conditional, logistic regression model. For the chi-square test and the multivariate logistic regression model, a two-tailed P-value less than .05 was considered to denote statistical signifi- cance. The statistical software SPSS, version 15.0 (SPPS Inc., Chicago, IL) was employed. One hundred forty-six patients were included in the study; 21 (14.4%) were neglected, and 125 (85.6%) were supported. Patient characteristics are shown in the Table 1. No differences were observed between the two groups for most of the factors studied. Only five characteristics had statistically significant differences between neglected and supported patients in the univariate analysis: metabolic disease (P 5.03), marital status (P 5.009), lack of offspring (P 5.005), and living in good socioeconomic conditions (Po.001). In the multi- variate logistic regression model, presence of spouse (P 5.02, Exp (B) 5 0.22) and lack of offspring (P 5.04, Exp (B) 5 3.97) remained independent statistically signifi- cant factors. This study shows that elderly Cretan patients who had a spouse and offspring were more likely to be supported during hospitalization. This finding is not surprising be- cause of the strong family bonds of Cretan society, which is dominated by traditions and in which negligence of elderly people is cause for social stigma. 5,7,8 Nevertheless, taking into account the social structure, the number of neglected patients should be smaller. The changing nature of the traditional societies, in which many young people, espe- cially women, work outside the home may explain this phenomenon. 8 The relationship between marital status, health and mortality has been extensively evaluated. Studies suggest that married people make less use of healthcare facilities 9 and that being married protects against mortality in later life, although mainly in men. 10 The University Hospital of Heraklion serves Crete as tertiary health center for the island. Patients cared for in this hospital represent all social levels and reflect local society. Most Cretans tend to remain in local towns and villages and do not seek treatment in urban centers of mainland Greece. Family bonds are strong and strict, and family networks are extensive. 7,8 Support of elderly people 1302 LETTERS TO THE EDITOR JULY 2009–VOL. 57, NO. 7 JAGS