The organisation of diabetes care for persons with type 2 diabetes mellitus : A nationwide cross-sectional study in Swedish primary health care

Abstract: Aim To extend the current knowledge about changes in Swedish diabetes care from 2006 to 2013 (Paper I), and to investigate the association between personnel resources and organisational features of primary health care centres (PHCCs) and the individual level of glycated haemoglobin (HbA1c) among persons with type 2 diabetes mellitus (T2DM) (Paper II).Methods The longitudinal cross-sectional study described in Paper I included a total of 684 (74.3%) of eligible 921 PHCCs in 2006 and 880 (76.4%) of eligible 1152 PHCCs in 2013. The Swedish National Survey of the Quality and Organisation of Diabetes Care in Primary Healthcare (Swed-QOP) questionnaire was sent to PHCC managers in 2007 and 2014. In Paper II, persons with T2DM (n = 230 958) attending 846 PHCCs were included. The PHCC-level data for 2013 were obtained from the Swed-QOP questionnaire and the individual-level data for 2013 were obtained from the National Diabetes Register. These data were linked to national registers containing individual-level data on socio-economic status and comorbidities.Results From 2006 to 2013, the median list size of PHCCs decreased while the median number of persons with T2DM at the PHCC increased (all P < 0.001). The median number of whole time equivalent registered nurses (RNs) per 500 persons with T2DM increased from 0.64 in 2006 to 0.79 in 2013 (P < 0.001). Compared with 2006, 30% percentage points more PHCCs (17.0% in 2006 and 47.5% in 2013) had an RN with at least 16 European Credit Transfer and Accumulation System (ECTS) credits in diabetes-specific education in 2013 (P < 0.001). Access to an in-house psychologist increased from 25.9% in 2006 to 47.1% in 2013 (P < 0.001). No change was found regarding in-house access to dietitians, chiropodists, social workers or physiotherapists. Observed changes in the organisational features of PHCCs from 2006 to 2013 included increased frequency of medical check-ups, use of call–recall systems and use of systems for checking that persons with T2DM participated in annual visits to general practitioners (GPs) and RNs (all P < 0.05). Providing group education remained scarce in both years (Paper I). After adjusting for several important confounders, personnel resources associated with decreasing individual HbA1c levels were the number of ECTS credits in diabetes-specific education among RNs (P < 0.001) and the duration of regular visits to RNs (P < 0.001). However, the opposite effect (i.e., increased HbA1c levels) was found for the duration of visits to GPs (P < 0.001). Organisational features associated with decreased individual HbA1c levels were PHCCs having a diabetes team (P < 0.01) and providing group education (P < 0.01) (Paper II).Conclusions The findings presented in this thesis add important knowledge on evaluating changes in diabetes care in Sweden. Some findings can be interpreted as initiatives to focus on increasing the time RNs devote to caring for persons with T2DM, and in ensuring RNs undergo diabetes-specific education. Furthermore, factors such as diabetes-specific education, diabetes teams and group education were associated with lower individual HbA1c levels. These findings could stimulate decision-makers to prioritise these factors for offering high-quality, equitable care to persons with diabetes.