This is a two-wave cross-national and longitudinal study. Data were collected in both Scotland and Italy, and the procedure differed slightly between the two countries. SCOTTISH PROCEDURE Five General Practitioner (GP) surgeries situated across Scotland posted invitations to participate in a two-wave longitudinal study to all their patients for whom the study was deemed suitable (those over 18 without terminal illnesses or conditions such as dementia or Alzheimer’s disease). No incentives were provided. Interested patients returned a reply slip included with the invitation. They received a questionnaire, to be completed and returned to the researchers. A member of the research team then visited the GP surgeries around 2 months after the completed questionnaires were received, and extracted health data for each participant from the databases. Approximately two years later, all the participants who had completed the first questionnaire received the second questionnaire via post, to be completed and returned to the researchers. Approximately 2 months after receiving the follow up questionnaires, a member of the research team again visited the GP surgeries in order to extract health data for each participant from the databases. ITALIAN PROCEDURE GPs working in five different surgeries (all based in rural areas in Tuscany, and all members of the same medical cooperative) asked patients visiting the surgeries if they wished to participate in a two-wave longitudinal study. Patients were only invited to participate if they were over 18 and did not possess a terminal illnesses or a condition such as dementia or Alzheimer’s disease. Eligible patients were instructed to complete the first questionnaire either in the surgery waiting room or at home. Patients who agreed to participate handed their completed questionnaires into their GP surgery. Around two months after receiving the completed questionnaires, the GPs extracted health data for each participant from their databases. Approximately two years later, all the participants who had completed the first questionnaire and were visiting the surgery were approached again and invited to complete a second questionnaire, either in the surgery waiting room or at home. This second questionnaire was also completed and handed into the GP surgery. Once again, health data for each participant was extracted from the databases by the GPs around two months after receiving the completed second questionnaires. The aim of the research is to explore the interplay between group identification and health, overcoming the limitations of previous studies.This involves examining the extent to which group identification influences different aspects of mental and physical health. It is predicted that group identification will have more impact on health than simple physical contact with group members. Moreover, the effects of group identification on health will be compared with the effects exerted by other factors known to predict health, such as socio-economic status and health-related behaviours. The degree to which the effects of group identification on health are independent from the effects of these other factors will also be considered. The prediction that the effects of group identification on health are mediated by perceived social support (ie, the extent to which group members feel they can rely on each other for assistance) will also be tested. To achieve these aims, the proposed research will: (1) test a large sample of participants (N = ~16,000), drawn from two nations (~8,000 from Scotland and ~8,000 from Italy); (2) use a longitudinal design (two waves across a two-year period: 2012-2014); (3) use an extended range of physical and mental health indicators.

Data were collected partly via a self-completion questionnaire and partly by scanning medical databases.