BackgroundChronic widespread pain (CWP) is common and associated with prominent negative consequences. The aim of this study was to assess the prevalence of persistent CWP in an 11-year prospective cohort study in the general population, and to examine anxiety, depression, alcohol use, poor sleep, body mass index (BMI) and chronic disease, along with demographic, lifestyle and other health-related variables as possible predictors for the assumed CWP persistence.MethodsCWP was defined as having pain at three or more predefined sites (involving the trunk and upper and lower limbs) for at least three months in the last year. We used a Norwegian general population cohort of 28,367 individuals who responded to both the second (1995–1997) and the third (2006–2008) waves of the Nord-Trøndelag Health Study (HUNT2 and HUNT3, respectively). Data were analysed with logistic regression models.ResultsCWP prevalence in HUNT2 was 17%. Of those reporting CWP in HUNT2, 53% still reported CWP at follow-up in HUNT3. Adjusted analyses revealed that depression and alcohol consumption were not substantially associated with the 11-year prospective CWP outcome. Poor sleep, obesity and chronic disease predicted persistent CWP, and being male and/or 60 years or older was protective.ConclusionsThis cohort study revealed that nearly half of the participants with baseline CWP resolved from CWP 11 years later. Among those whose CWP did not resolve, obesity, sleeping problems and chronic disease predicted CWP persistence, while aging and male sex was protective. Anxiety, mixed anxiety and depression, former smoking, and overweight were weakly associated, while depression, moderate exercise, and alcohol use were not associated with persistent CWP.

Assessing activation before treatment might be useful for scheduling the delivery of mental health services as well as evaluating educational interventions aimed at improving patient engagement in mental health.

Behavior and symptoms seen in 48 24-h periods preceding violent episodes and 93 control observations were studied. Six behaviors were more common before violence: confusion, irritability, boisterousness, physical threats, verbal threats and attacks on objects. A logistic regression equation based on these behaviors in a randomized half of the observations predicted the occurrence of subsequent violence in 92.1% of the other half of the sample without any false positives, giving a sensitivity of 81.3% and a specificity of 100%. It is concluded that mentally ill people display the same behaviour before violent acts as we would expect in people without such disorder. The potential for short-term prediction of violence seems good.