primary care diabetes 7 ( 2 0 1 3 ) 71–89 83 P25 Case note survey of T2D patients prescribed incretin therapies according to current recommendations in clinical practice M. Evans 1 , P. McEwan 2 , R. O’Shea 3 , L. George 1 , C. Svendsen 4 , A. Clarke 1 1 University Hospital Llandough, Llandough, Penarth, South Glam- organ, UK 2 SwanseaUniversity, Swansea, UK 3 University of Wales College of Medicine, Heath Park, Cardiff, UK 4 Novo Nordisk A/S, Bagsvaerd, Denmark Aims: To assess the clinical and cost-effectiveness and patient preference for incretin-based therapies when initiated according to National Institute for Health and Clinical Excel- lence (NICE) recommendations in UK clinical practice. Methods: In a retrospective chart audit, anonymised data were collected for patients receiving incretin-based ther- apy in clinical practice according to NICE recommendations. Parameters assessed included HbA1c, weight, the proportion of patients achieving NICE treatment continuation criteria, adverse events, and treatment discontinuation. Based on observed treatment effects, drug cost-effectiveness was also assessed. Treatment preference for a dipeptidyl peptidase-4 (DPP-4) inhibitor or glucagon-like peptide-1 receptor agonist (GLP-1RA) was assessed prospectively. Results: 1114 patients were followed up for a median of 48 weeks (256 received liraglutide, 148 exenatide, 710 DPP-4 inhibitor). Liraglutide reduced HbA1c significantly more vs. exenatide or DPP-4 inhibitor (both p < 0.05). Weight changes were similar for both GLP-1RAs but were significantly greater vs. DPP-4 inhibitors (both p < 0.05). NICE treatment contin- uation criteria were met by 32% and 24% of liraglutide 1.2 mg- and exenatide-treated patients (≥1% HbA1c reduc- tion, ≥3% weight loss) and 61% of DPP-4-inhibitor-treated patients (≥0.5% HbA1c reduction). Life-years/patient gained were 0.12, 0.08, and 0.07, and costs/quality-adjusted life-year were £16,505, £16,648, and £20,661 for liraglutide, exenatide and DPP-4 inhibitors, respectively. Most patients preferred the GLP-1RA profile, with these patients having higher BMI, HbA1c and diabetes duration than those preferring the DPP-4 inhibitor profile. Conclusions: Liraglutide provided the greatest reductions in HbA1c and bodyweight, with a more cost-effective pro- file than exenatide or DPP-4 inhibitors. Patients with more advanced disease preferred GLP-1RAs vs. DPP-4 inhibitors. http://dx.doi.org/10.1016/j.pcd.2012.10.068 P26 Cancer as a leading cause of mortality in a cohort of newly diagnosed type 2 diabetic patients N. Soldevila-Bacardit, J. Torras-Borrell, M.I. Fernández- Sanmartín, M. Mata-Cases Barcelona Family and Community Medicine Teaching Unit, Catalan Institute of Health, Barcelona, Spain Aims: To describe rates and causes of mortality in a cohort of newly diagnosed type 2 diabetes mellitus (T2DM) patients in primary care. Methods: Observational cohort study of 598 T2DM patients diagnosed in an urban primary care centre from 1991 to 2000 and followed up until July 2011. Patients without a previ- ous glycaemic test in the three years before diagnosis were excluded to assure they were at the beginning of the disease. Causes of death were obtained from electronic clinical records and validated in the mortality register of the Spanish National Statistics Institute. Mortality rates per 1000 patients/year for total, cardiovascular, cancer and other causes by sex and age at diagnosis were calculated. Kaplan-Meier survival curves were made and compared using the log-rank test. Results: 469 patients fulfilled the inclusion criteria (mean age [SD]: 60.4 [10.7] years; 53.9% women). 146 (31.1%) patients died: 72 men and 74 women during a median follow-up of 13 years. Mortality rates per 1,000 were 26.98 (95%CI 21.11–33.97) in men and 22.55 (17.71–28.31) in women. The cancer mortality rate (7.39; 5.37–9.93) was greater than the car- diovascular rate (6.22; 4.38–8.57), and significantly different by sex for can-cer (11.61; 7.89–16.49 in men and 3.96; 2.11–6.78 in women; p = 0.001). The most frequent were lung cancer (12.5% in men and 0% in women; p < 0.01) and colorectal can- cer (8.3% in men and 2.7% in women, p = ns). Kaplan–Meier survival curves showed that a history of smoking (p = 0.007) and mean HbA1c < 7% during follow-up (p = 0.027) were signif- icantly related to lower survival in men, but not in women. Mean survival time was lower in men than in women in the < 65 age group (p = 0.05). Conclusions: Cancer was the first cause of mortality in men and cardiovascular disease was the first cause of mortality in women. Only one quarter of patients died due to a cardiovas- cular cause. http://dx.doi.org/10.1016/j.pcd.2012.10.069 P27 A descriptive study about blood glucose measurements and factors affecting diabetes in women with two different socio- economic profiles in Istanbul Z. Polat, S. Yolcu, S. Bayram, I.Y. Keles Acibadem University School of Medicine, Istanbul, Turkey Aim: Screening and metabolic and anthropometric mea- surements in two groups of women who applied to family health-care centres (FHHCs) for any reason (without diag- nosis of diabetes). The FHHCs involved deal with people from different socioeconomic backgrounds. We compared these metabolic syndrome parameters between two groups of women. Design and method: The study included 100 women between 18-50 years of age who applied to the Gülsuyu (socio- economically low) and Erenköy (socio-economically high) FHHCs who had not been diagnosed with diabetes. Of the 147 people interviewed, 28 refused to take part in the study and 19 were not included in data analysis. Those willing to partic- ipate signed an informed consent form. Spot capillary blood glucose, BMI, body fat ratio and anthropometric parameters were measured. A face-to-face questionnaire was applied in order to gauge their socioeconomic status (p < 0,05).