Online-Only Appendix
Additional Project information
The Styrian Diabetes Type 2 Education project was started following an agreement among the local Medical Association, the main regional health insurance fund (Gebietskrankenkasse), the Austrian Forum for Quality Management Systems in Diabetes Care (FQSD), the Austrian association of diabetes educators and the Styrian provincial government. Styria, with a population of 1.2 million, is a rural, mountainous province with small cities and widespread villages. The training courses for educating staff (diabetes educators and physicians) are held by internists specialised in diabetology and contain information about the content, objectives, administration and organisation of the programme. These courses contain information about the content, the objectives and the administration and organisation of the programme. There is a teaching unit on prevention of secondary complications, including the demonstration of a correct foot examination. A refresher course is recommended to be attended 12 months after the two day postgraduate DTTP training course. Diabetes educators (nurses and dietologists), who assist by giving single lessons during the DTTP, participated in a special half-day theoretical training programme to familiarize with the structure of the DTTP. The doctor’s fees are funded by the local insurances, financed by compulsory health insurance and amount to US$ 870 for one course (4 sessions/ 9 teaching units with 3 to 12 participants). For the follow-up course (2 hours) another US$ 183 is paid. These sums are directly transferred to the physician after each course and follow-up. A teaching unit (45 minutes) done by a diabetes educator is paid with the sum of US$ 54, which is subtracted from the total amount paid to the physician. In this system patient teaching is conducted by physicians and diabetes educators, whereas physicians have to perform at least 3 out of 9 teaching units themselves. In most of the cases the other 6 units were conducted by diabetes educators. The completion of the data collection sheet (www.fqsd.at/export/sites/fqsd/de/werkzeuge/FQSD-BIS.pdf) after the training programme and at follow up is rewarded another US$ 19 each.
The DTTP used in our study was developed and evaluated in Germany and has demonstrated efficacy in various randomized controlled trials in many health care systems (9), therefore a control or comparison group in our study was not necessary. The results of one study performed in a rural area in Styria were used as the basis for the final agreement on regional implementation of the programme (7). Briefly, the DTTP consists of 4 sessions (theoretical, practical and experimental parts; 90-120 min each, 3 to 12 participants) and covers 9 education areas: basic diabetes information, self-monitoring, medication and hypoglycaemia, diet, foot care, physical activity, sick day rules and late complications. A teaching set is used, consisting of a series of flip-charts, photographs of foodstuff, memory cards and literature. Recruitment for the DTTP is undertaken by physicians during their consultations. A criterion for inclusion is the diagnosis of non insulin treated type 2 diabetes. There is no exclusion criterion, like for example advanced age or decreased mental status, although the decision for proposing a training course to the patient is taken on a subjective basis by the treating physician. Patients willing to participate in the programme were free to bring the partner with them. The only modifications to the original programme were the introduction of a permanent quality management and refresher training for patients and physicians after one year.
Statistical Analysis
Between-group comparison of continuous data was performed by means of the Wilcoxon test and comparison of categorical data with the χ2 test. Within-group changes were analysed by means of one sample Wilcoxon and McNemmar tests for continuous and binary data, respectively. Statistical analyses were performed with R software (R Development Core Team: A Language and Environment for Statistical Computing, 2005. Available from http://www.R-project.org, accessed 10 March 2006). A two-sided p value less than 0.05 was considered statistically significant.
Baseline characteristics
Overall baseline characteristics and comparisons of participants with and without follow up after 1 year are given in the table below. Attendees of a follow-up assessment (48%) had a significantly longer duration of diabetes, smoked less often, and had higher HbA1c levels, lower blood pressure, somewhat worse lipid profile and higher prevalence of myocardial infarct and neuropathy in comparison to non-attendees at baseline.
Table 1: Baseline comparisons of participants with and without follow up after 1 year
(n= 2274) / Follow up
(n= 2122) / Overall
(n= 4396) / P value
Age (years) / 63.3±11.2 / 64.2±10.2 / 63.8±10.7 / 0.03
Duration of diabetes (years) / 4.5±6.1 / 5.5±6.4 / 5.0±6.2 / < 0.001
Gender (female) / 54.0 / 56.1 / 55.0 / 0.17
Smokers (%) / 14.0 / 11.7 / 12.9 / 0.03
BMI (kg/m2) / 29.8± 5.4 / 29.6 ± 4.8 / 29.7± 5.1 / 0.55
HbA1c (%) / 7.5 ± 1.6 / 7.6±1.6 / 7.6±1.6 / 0.001
Systolic BP (mm Hg) / 143.2 ± 19.4 / 142.1± 18.2 / 142.7 ± 18.8 / 0.03
Diastolic BP (mm Hg) / 83.2 ± 10.3 / 82.2 ± 9.9 / 82.7 ± 10.1 / 0.01
Creatinin (μmol/l) / 88.3 ± 46.8 / 85.7 ± 26.1 / 87.0 ± 38.2 / 0.05
Cholesterol (mmol/l) / 5.62 ± 1.28 / 5.61 ± 1.13 / 5.61±1.21 / 0.62
LDL (mmol/l) / 3.38 ± 1.04 / 3.44 ± 1.0 / 3.41 ± 1.02 / 0.06
HDL (mmol/l) / 1.30 ± 0.42 / 1.27 ± 0.43 / 1.28 ±0.42 / 0.04
Triglycerides (mmol/l) / 2.19 ± 1.69 / 2.18 ± 1.57 / 2.18 ± 1.63 / 0.77
Complications at Baseline
Myocardial infarct / 7.0 / 9.2 / 8.1 / 0.01
Stroke / 5.0 / 5.7 / 5.3 / 0.32
Angina Pectoris / 14.0 / 15.9 / 14.9 / 0.08
Painful Neuropathy / 17.9 / 22.4 / 20.1 / < 0.001
Microalbuminuria
Macroalbuminuria
ESRD / 16.1
15.0
0.8 / 18.6
12.2
0.8 / 17.3
13.7
0.7 / 0.18
Non prolif. Retinopathy
Proliferative Retinopathy
Blindness / 10.6
0.3
1.7 / 10.7
0.8
2.4 / 10.6
0.6
2.0 / 0.34
Acute foot ulcus / 1.7 / 1.5 / 1.5 / 0.80
Data are mean ± SD or %, BP: blood pressur, ESRD: end stage renal disease
Regional differences in training frequencies:
3898 participants (89%) have been trained by physicians in private practice and 498 (11%) in outpatient clinics. In order to increase the number of courses in remote areas, a mobile team was installed. There were a significantly lower number of courses held in these remote areas (see the figure below), although this is biased by the mobile education team, for which each training is accounted to the capital Graz, regardless of the place of teaching.
Figure 1: Number of courses per 1000 inhabitants in each Styrian district (16 districts including the capital Graz)
Relative HbA1c shift
International guidelines propose HbA1c values around 7-7.5% as a cut-off level for reduced risk for micro- and macrovascular complications (Grüsser M, Bott U, Scholz V, Kronsbein P, Jörgens V: The introduction of a structured treatment and teaching program for type 2 diabetes in general practices [in german]. Diab Stoffw 1:229-234, 1992 and American Diabetes Association: Standards of Medical Care in Diabetes. Diabetes Care. 29:S4–S42, 2006). As shown in figure 2, the proportion of participants with HbA1c levels below 7% increased from 43% to 49% (p<0.001), so there could be a benefit in terms of a reduction of diabetes related complications. Contrarily the proportion of patients with HbA1c levels higher than 9%, which can be considered poorly controlled, decreased from 17% to 9% after the intervention.
Figure 2: Percentage of patients within each HbA1c category. Baseline data are represented with rectangles and black lines, and follow-up data with triangles and dotted lines
5
Treatment for arterial hypertension and blood lipids
Whilst a treatment for arterial hypertension was being prescribed for 66% of the patients prior to the course and for 62% at follow-up (p<0.001), the proportion of patients with blood pressure levels below 140/90 mm Hg increased from 56% to 60% (p=0.01). The fact that the proportion of patients with blood pressure levels below 140/90 mm Hg increased although there was a decrease in drug prescription for arterial hypertension is probably related to a higher compliance in drug intake after the training course.
Lipid lowering treatment was prescribed in 40% and 41% (p=0.30) of patients at baseline and follow-up, respectively, and the proportion of patients with cholesterol values below 5.2 mmol/l (200 mg/dl) increased from 37% to 42% (p<0.001).
7