Depression in the Lundby Study 1947-1997 Incidence, course and risk-factors

Detta är en avhandling från Department of Clinical Sciences, Lund University

Sammanfattning: Abstract The aims of this thesis were to describe methodological problems, incidence and course of depressive disorders as well as possible risk factors for depression within the framework of the Lundby Study. The Lundby population was investigated in 1947, 1957, 1972 and 1997. In 1947 the population consisted of 2550 subjects and in 1957, 1013 were added. The total population consists of 3563 subjects. Attrition rate for the field-investigations, 1947, 1957 and 1972 was 1-2% and in 1997 6%. During the follow-up period the Lundby study had encountered many methodological problems such as changing diagnostic systems, recall bias, low interrater-reliability over time, migration and attrition. Low attrition rate over 50 years and reasonable diagnostic uniformity justify comparisons over time. Incidence of depression was compared for the two time periods 1947-1972 and 1972-1997. Females had higher incidence rates than males in both periods. In the period 1972-1997 the average annual incidence rate was lower for females and tended to be lower for males as compared with 1947-1972. The course of depression was followed for 344 subjects that had experienced an episode of first incidence depression during follow-up. The recurrence rate was about 40 % . Changes to other diagnoses were registered in 21% of the this sample, alcohol disorders in 7 % and bipolar disorders in 2 %. Five percent committed suicide; male gender and severity of depression were significant risk factors. Risk factors for the whole sample and for the genders separately were analyzed for the cohorts1947 and 1957. For the 1947 cohort the personality trait nervous/tense was a risk factor for both genders in univariate analyses as well as in the multivariate analyses. Prior anxiety disorders, tiredness disorder, alcohol disorder and subvalidity appeared to be risk factors, especially for males in univariate analyses. In the multivariate model a risk factor for males was child neurosis and for females the personality trait abnormal/antisocial. In univariate and multivariate analyses of the 1957 the cohort, the risk factors anxiety disorders, alcohol disorders and childneurosis were risk factors for males. For females the personality traits nervous/tense, abnormal/antisocial, tired/distracted, easily hurt and anxiety disorders were risk factors. In multivariate analyses the personality traits nervous/tense and abnormal/antisocial turned out as risk-factors for females.