The impact of screening for Type 2 diabetes in siblings of patients with established diabetes

Background Targeted screening for Type 2 diabetes has been proposed as a method of identifying people with, or at risk of, the disease in order to implement preventive care. Aim To assess the changes in anxiety, well-being, and cognitions associated with screening for Type 2 diabetes in people at increased risk of diabetes after 1 year; and to identify potential predictors of increased anxiety and lower well being over this period. Design and setting One-year follow up of a cohort of non-diabetic siblings of those with Type 2 diabetes registered with general practitioners in Oxfordshire and Northamptonshire who were identified for participation in a programme to undertake sib-pair genetic analysis. Methods Potential families were selected through identification of people with Type 2 diabetes. Family members aged 35-74 years, who did not have diabetes and who were willing to participate in the study, completed measures before receiving the results of a fasting plasma glucose test, and 1 year later. Measures included the Spielberger State Anxiety Scale-short form (SSAI-SF), the Well-Being Questionnaire 12 item scale (WBQ-12), and measures of cognitions about developing diabetes. The Health Anxiety Inventory (HAI) was completed before receiving the results of the tests to assess its relationship with anxiety at 1 year. Results A cohort of 431 individuals was identified, of whom 328 (76%) returned fully completed WBQ scales both initially and at 1 year. State anxiety measured with the SSAI-SF fell from 34.5 (95% CI 33.4-35.6) to 32.3 (31.2-33.4) at 1 year (P < 0.0001). Mean WBQ-12 scores rose (i.e. improved) from 26.8 (26.0-27.4) to 27.4 (26.7-28.1) (P = 0.008). SSAI-SF and WBQ-12 mean scores over 1 year did not differ between participants receiving a normal or an at-risk test result. However, those receiving an at-risk test result were more likely, at 1 year, to consider themselves at increased likelihood of developing diabetes (P < 0.001) and to report thinking about what it would be like to develop diabetes (P = 0.006). A score in the upper tertile of the initial HAI was associated with an increased level of anxiety at 1 year (adjusted odds ratio 2.0, 95% CI 1.2-3.4, P = 0.006). Conclusions There is no evidence that an 'at-risk' test result leads to sustained anxiety or reduced well being at 1 year compared with those receiving a normal test result. However, further studies are required to clarify levels of anxiety prior to screening to assess the overall impact of the process.