OBJECTIVE:To estimate the prevalence and associated factors of depressive symptoms in the elderly.
METHODS:Cross-sectional population based epidemiological study (The EpiFloripa Elderly survey) was carried out in two stage clusters, census tracts and households, with1.656 elderly individuals in Florianópolis, SC. The prevalence of depressive symptoms (outcome) was obtained using Geriatric Depression Scale (GDS-15) and its associations with socio-demographic, health, behavioral and social variables were assessed. Crude and adjusted prevalence ratios (PR) with 95% confi dence intervals were calculated using Poisson regression.
RESULTS:Depressive symptoms were observed in 23.9% of the elderly individuals (95%CI 21.84;26.01). In the fi nal model, depressive symptoms were associated with: 5 to 8 years of schooling (PR = 1.50, 95%CI 1.08;2.08); one to four years of schooling (PR = 1.62, 95%CI 1.18;2.23) and no schooling (PR = 2.11, 95%CI 1.46;3.05); being in a worse fi nancial condition than at the age of 50 (PR = 1.33, 95%CI 1.02;1.74); cognitive impairment (PR = 1.45, 95%CI 1.21;1.75); perceiving their health to be regular (PR = 1.95, 95%CI 1.47;2.60) or poor (PR = 2.64, 95%CI 1.82;3.83); functional dependence (PR =1.83, 95%CI 1.43;2.33) and chronic pain (PR = 1.35, 95%CI 1.10;1.67). Elderly people often suffer from depression and it is highlighted as one of the most common and signifi cant psychiatric problems. 2,8 According to a systematic review, prevalence worldwide varies between 0.9% and 9.4% in those living in the community and 14% to 42% in those living in a care home.8 Brazilian epidemiological studies carried out with the elderly show that the prevalence of depressive symptoms varies between 195 and 34% in the different regions of the country. 4,5,10,15 The high prevalence of depression requires the attention of health care professionals and managers, as this disease increases the likelihood of functional incapacity in the elderly 16 and constitutes a signifi cant public health problem.Among the factors linked to depression, the following stand out: being female, insomia, 6,10 and a predominance of negative exchanges in relationships with family and friends. 23 These factors, especially those that can be modifi ed, need to be further explored, taking into account cultural differences between countries and regional differences within those countries, as in Brazil.Identifying the factors linked to the presence of depressive symptoms may help prevent depression among the elderly 3 and give an opportunity to non-pharmacological treatment strategies. The following were assumed in calculating the general sample for the EpiFloripa Elderly Survey: a population of 44,460, b unknown outcome prevalence (50%), sample error equal to four percentage points, 95% confi dence level and design effect equal to two. An additional 20% was added to cover the predicted losses, and another 15% in order to control confounding factors, giving a minimum sample size of 1,599 individuals.The selection process for the sample w...