Differential shared care for diabetes: does it provide the optimal partition between primary and specialist care?

Aims To establish whether a system of differential shared care between gene ral practitioners and specialists is compatible with patients receiving the level of care they require. Methods We sought to trace 200 shared care patients whose care had been kep t at the general practitioner level after initial referral and compared the m with a group of patients who had been re-referred to the Royal Prince Alf red Hospital Diabetes Centre for specialist review. Results There were no significant differences in glycaemic, blood pressure and lipid levels of returned and non-returned patients at initial assessmen t. However, non-returned patients were less likely to have a history of mac rovascular disease or risk factor (adjusted odds ratio (OR) 0.4; 95% confid ence interval (CI) 0.2-0.6). Their referral letter was also more likely to emphasize their type and/or duration of diabetes (adjusted OR 4.6; 95% CI 2 .5-8.4). Nearly half (47.1%) of the non-returned group changed their doctor in the years following their initial specialist review, increasing their l ikelihood of not being re-referred five-fold (adjusted OR 5.0; 95% CI 2.9-8 .8). At initial assessment, non-returned patients were given less treatment recommendations (adjusted OR 0.5; 95% CI 0.3-0.7). Doctors registered with the Diabetes Shared Care Programme referred more patients than their non-s hared care counterparts. However, a higher proportion of these doctors (52. 5% vs. 21.3%; chi (2) = 16.5, 1 d.f., P = 0.00005) were selective in whom t hey re-referred. Conclusion Differential shared care encourages appropriate referral to spec ialist services, without compromise to standards of care.