Erratum to: Soc Psychiatry Psychiatr Epidemiol DOI 10.1007/s00127-012-0515-8

Unfortunately, in the article’s abstract the critical word ‘not’ is missing from the results statement: “Inserting measures of formal and informal social participation and perceived social support into the logistic regression models did significantly reduce these odds ratios.”

Please find the correct abstract below:

Abstract

Purpose

Loneliness can affect people at any time and for some it can be an overwhelming feeling leading to negative thoughts and feelings. The current study, based on the Adult Psychiatric Morbidity Survey in England, 2007, quantified the association of loneliness with a range of specific mental disorders and tested whether the relationship was influenced by formal and informal social participation and perceived social support.

Methods

Using a random probability sample design, 7,461 adults were interviewed in a cross-sectional national survey in England in 2007. Common Mental Disorders were assessed using the revised Clinical Interview Schedule; the diagnosis of psychosis was based on the administration of the Schedules of the Clinical Assessment of Neuropsychiatry, while loneliness was derived from an item in the Social Functioning Questionnaire.

Results

Feelings of loneliness were more prevalent in women (OR = 1.34, 95 % CI 1.20–1.50, P < 0.001) as well as in those who were single (OR = 2.24, 95 % CI 1.96–2.55, P < 0.001), widowed, divorced or separated (OR = 2.78, 95 % CI 2.38–3.23, P < 0.001), economically inactive (OR = 1.24, 95 % CI 1.11–1.44, P = 0.007), living in rented accommodation (OR = 1.73, 95 % CI 1.53–1.95, P < 0.001) or in debt (OR = 2.47, 95 % CI 2.07–1.50, P < 0.001). Loneliness was associated with all mental disorders, especially depression (OR = 10.85, 95 % CI 7.41–15.94, P < 0.001), phobia (OR = 11.66, 95 %CI 7.01–19.39, P < 0.001) and OCD (OR = 9.78, 95 % CI 5.68–16.86, P < 0.001). Inserting measures of formal and informal social participation and perceived social support into the logistic regression models did not significantly reduce these odds ratios.

Conclusion

Increasing social support and opportunities for social interaction may be less beneficial than other strategies emphasising the importance of addressing maladaptive social cognition as an intervention for loneliness.