The effectiveness was examined of inactivity and weight loss as criteria to identify a frail subgroup within independently living elderly persons participating in the SENECA study (Survey in Europe on Nutrition and the Elderly, a Concerted Action). Eight-hundred-forty-nine participants (aged 75 to 80) from nine countries were classified in four subgroups: 1) inactive elderly (lowest tertile activity score: n = 204); 2) weight losing elderly (lowest quintile: * 6.3% weight loss over 4-5 years: n = 108); 3) both inactive and weight losing (n = 54); 4) neither inactive nor weight losing: reference (n = 483). Differences in health, physical functioning and nutritional characteristics between groups 1, 2 and 3 respectively, and the reference group were evaluated. Compared to the weight-stable, active reference group, both inactive, weight losing (group 3) and inactive subjects (group 1) reported significantly more chronic diseases (2.2 and 1.8 vs. 1.1), disabilities (81 and 80 vs. 43%), use of medications (both 2.3 vs. 1.1) and care services (26 and 21 vs. 6%), and a lower self-rated health (2.8 and 3.1 vs. 3.8), relative health (1.9 and 2.1 vs. 2.6), and physical performance score (17 and 18 vs. 22). In addition, inadequate micronutrient intake and biochemical deficiencies were more prevalent. Weight-losing elderly were not significantly different from the reference group with respect to these characteristics. Therefore, physical inactivity alone or in combination with weight loss seems to be a practical and inexpensive screening criterion for identifying a subgroup of elderly with less favourable health and nutritional characteristics and poorer physical functioning among non-institutionalised elderly.