To determine whether the effect of multimorbidity on time to mortality is modified by level of social support and loneliness in a representative sample of 2113 participants aged 60 +.Methods:
Vital status was ascertained through national registers or by asking participants' relatives. Baseline variables included number of illnesses, self-perceived social support (Oslo social support scale) and loneliness (UCLA loneliness scale). Kaplan-Meier survival curves were used to estimate the time to death by multimorbidity, social support and loneliness. Adjusted cox proportional hazards regression models were conducted to explore interactions between multimorbidity and social support and loneliness.Results:
Multimorbidity was associated with low probability of survival, whereas high loneliness and low social support were not related with time to death. Only the interaction multimorbidity * social support was significant. Participants with low social support and 2 chronic diseases, compared with none, presented lower probability of survival (HR = 2.43, 95%CI = 1.14–5.18, p < 0.05), whereas the effect of multimorbidity, in comparison with not having chronic conditions, did not affect mortality if participants had high social support. For participants with low social support, there were no differences between having one, two or more than two diseases. When there is high social support, the probability of death is significantly lower if one or two chronic diseases are present, compared with more than two.Discussion:
These findings indicate that having a supportive social environment increases the survival of people with physical illnesses, especially those with one or two. For those with more than two illnesses, survival remains unchanged regardless of the level of social support and other protective factors should be explored in future research. Geriatric health professionals are encouraged to evaluate social relationships and stimulate support given by relatives, friends or neighbors.