The predictors of multiple somatic symptoms in a population sample: The Lifelines cohort study

Abstract 
Objective:  Multiple bodily symptoms predict poor health status, high healthcare use and onset of functional somatic syndromes. This study aimed to identify the predictors of multiple somatic symptoms in a population sample. 
Methods: The prospective, population-based Lifelines cohort study included 92,539 adults who were followed up for a mean of 2.4 years. At baseline socio-demographic status, self-reported medical disorders, neuroticism, psychiatric disorders (assessed with MINI) and other variables including SCL-90 somatisation scale were assessed. The predictors of a high somatic symptom score at follow up were assessed using logistic regression which adjusted for baseline somatic symptom score (n=67,121). The relative predictive power of 7 dimensions was assessed using linear regression. 
Results: The strongest baseline variables which predicted a high SCL somatisation score at follow-up (n=20,664) were: baseline somatic symptoms score, fibromyalgia, irritable bowel syndrome, prolapsed intervertebral disc, frequent use of analgesia, agoraphobia, panic disorder and migraine. The first six of these plus GERD, negative health perception, GAD and few years of education predicted a persistent high somatisation score. In linear regression analysis summary scores of demographic features, medical, psychiatric and functional disorders, stress, neuroticism and current health status were all independently associated with SCL somatisation score at follow-up.  
Conclusions: The numerous predictors of multiple somatic symptoms probably underlie its ubiquitous predictive nature and may reflect numerous different aetiological pathways. The bi-directional relationship between functional somatic syndromes and multiple somatic symptoms is compatible with the notion that perception of bodily symptoms is a key component of both. 
Keywords:  Somatic symptoms, epidemiology, risk factors, somatisation.