This paper aims to investigate the causes that explain the discrepancies between the effects of social relationships on health and wellness of the elderly people. Several disciplines of health sciences have developed different theories to explain the evidence that confirm the positive effects of social relations. Furthermore, there is acumuative evidence confirming most of the predictions derived from its principles. However, the empirical evidence has not always confirmed these beneficial relationships and sometimes these evidences contradict some of the theoretical predictions. Even, it is not difficult to find reversed effects. In this paper we follow some of the approaches developed from Social Psychology which analyze the different effects of social relationships on the health of older people. Two types of aspects of social relations have served to this purpose. First, the structural aspects (i.e., frequency of intercourse). Second, qualitative-functional aspects of great tradition in estudies of quality of life and wellbeing of older people (i.e. social support). Following the Convoy Model, we measured perceived social support and frecuency of relationships in 168 spaniards, men and women (aged 62 years old and more). We analyze these discrepancies in the light of the underlying mechanisms.