The study examines in what way objective health-related variables interfere with psychic health and personality factors in explaining self-perception of health in the elderly. Two hundred and sixty-one patients aged 60 and older of an internal medicine hospital previously examined between 1994 and 1997 were once more contacted five years later. One hundred and sixty-four patients could not be included in the present investigation because of death, dementia, or severe physical illness. Of the 97 patients eligible for this second investigation, 74 agreed to participate. They were investigated extensively by means of psychometric scales and diagnostic interviews. A positive selection effect could be found for the sample of the present investigation with regard to age and health-related variables. Subjective evaluation of health correlated highly with the self-evaluation scales that recorded subjective well-being (life satisfaction, anxiety, and depression), and with the sense of coherence, but not substantially with objective health-related variables. A backward regression resulted in an adjusted R2 = 0.33 for the three retained variables "subjective physical complaints", "sense of coherence" and "self-evaluated depression" which rendered the same variance clarification of subjective health as did the model including all variables. Since the elderly represent the majority of patients treated in general hospitals and as subjective health and subjective physical complaints influence frequency of medical consultations and health care utilization, this is an important issue for consultation-liaison-psychiatry and health policy.