Sage Journals: Discover world-class research

Abstract

Background:

The number of patients with type 2 diabetes mellitus and diabetes mellitus–associated chronic kidney disease varies considerably between countries. Next to differences in genetic as well as life style risk factors, varying practices in medical care delivery might cause this diversity.

Method:

The PROVALID study recruited 4000 patients with type 2 diabetes mellitus at the primary level of healthcare in five European countries (Austria, Hungary, The Netherlands, Poland and Scotland). Baseline data were used to describe patient characteristics and compare the adherence to ADA (American Diabetes Association) and KDIGO (Kidney Disease: Improving Global Outcomes) guidelines with respect to metabolic and blood pressure control, use of renin–angiotensin system–blocking agents, statins and acetylsalicylic acid between the countries.

Results:

About 34.8% of the population had evidence of diabetes mellitus–associated chronic kidney disease. The median HbA1c level of the cohort was 6.8% (ranging from 6.5 in Poland to 7.0% in Scotland). Mean blood pressure was 136/79 (±17/10) and significantly higher in subjects with elevated albuminuria. These individuals also were more often treated with renin–angiotensin system–blocking agents (74.1% vs 84.6%), whereas the use of statins was driven by cardiovascular comorbidity. Acetylsalicylic acid was used in only 28.9% subjects. Despite similar cardiovascular comorbidities and renal function, the use of renin–angiotensin system–blocking agents varied significantly between the countries from 66.7% to 87.4%. An even higher variability was observed for patients >40 years of age using statins (39.8%–82.7%) and administration of acetylsalicylic acid in patients older than 50 years (5.2%–43.8%).

Conclusion:

Our study shows that medical practice in type 2 diabetes mellitus patients with and without renal disease is different in European countries. Longitudinal follow-up will reveal if this diversity affects clinical endpoints.

Keywords

Diabetes mellitus diabetic kidney disease ADA guideline adherence KDIGO guideline adherence

Background

The Global Burden of Disease Study ranked chronic kidney disease (CKD) 19th on the list of all causes of mortality¹ and the age-standardized death rate due to renal failure increased from 11.6 to 15.8/100,000 between 1990 and 2013. Diabetes mellitus–associated CKD (DKD) is a major contributor and the number of diabetic patients with end-stage renal disease (ESRD) in Europe is expected to rise annually by 3.2% at least until the year 2025.² Next to issues directly related to renal replacement therapy (RRT), comorbidities (like premature cardiovascular events, increased incidence of fractures, cognitive impairment, and anaemia) make DKD not only an important global health but also socio-economical problem.^3,4

In order to support clinical decision-making and healthcare policy planning, it is important to understand the epidemiology of CKD and identify factors associated with its incidence and progression, ideally not only in but also outside the setting of randomized controlled trials. Kramer at al.⁵ analysed the 2013 EDTA-ERA registry data on incident patients to RRT. On average, DKD was listed as the cause of ESRD in 24% (26 pmp), but interestingly the differences between countries were remarkable (from 6 in the Ukraine and 12 pmp in Iceland to 58 and 70 pmp in Greece and Portugal). Regional variability was observed also in non-dialysis-dependent CKD in Europe,⁶ the United States⁷ and China.⁸ Next to heterogeneities in study design, inclusion and/or exclusion criteria or laboratory methodology used for creatinine measurement,⁹ genetic background driven disease susceptibility as well as dietary habits can affect serum creatinine values and/or incidence or progression of the disease. Regional differences in risk factors like smoking, physical activity, socioeconomic status and birth weight also have to be considered. As far as initiation of RRT is concerned, patterns of referral to dialysis and/or transplantation may vary and finally local healthcare policies and/or clinical practice pattern (e.g. metabolic or blood pressure control) also might influence CKD/DKD incidence, prevalence and progression.

In order to assess the latter, we analysed baseline data from the PROVALID (PROspective cohort study in patients with T2DM for VALIDation of biomarkers) study, a prospective cohort study that recruited subjects with T2DM at the primary level of healthcare in Austria, Hungary, Poland, Scotland and The Netherlands. The aim of our study was to describe whether the presence or absence of renal disease as defined by the KDIGO (Kidney Disease: Improving Global Outcomes) stages¹⁰ affects major aspects of treatment in the total population and in the different participating countries. As a measure, we used the adherence pattern to ADA (American Diabetes Association) and KDIGO treatment guideline recommendations for metabolic, blood pressure and low-density lipoprotein (LDL) cholesterol (LDL-C) control as well as administration of aspirin.^11–13

Methods

Study design

A detailed description of the PROVALID study is presented elsewhere.¹⁴ In summary, PROVALID is a prospective cohort study of 4000 individuals with incident or prevalent T2DM (defined either as treatment with hypoglycaemic drugs or according to ADA guidelines). Subjects older than 18 years were recruited at the primary level of healthcare in five European countries (629 patients in Austria, 1420 in Hungary, 903 in The Netherlands, 537 in Poland and 511 in Scotland) and only those with active malignancy (defined as those currently receiving chemotherapy) or who declined to participate were excluded. Details on the recruitment procedure are described in the work by Eder et al.¹⁴ This article uses individual patient data obtained at the time of recruitment into the study (2011–2015). According to the KDIGO classification system¹⁰ DKD was considered absent in subjects with stages G1 and G2 [estimated glomerular filtration rate (eGFR) ≥60 mL/min/1.73 m²] and urinary albumin excretion <30 mg/g creatinine (A1, formerly called normoalbuminuria) in a morning spot urine sample.

Statistical analysis

Data are presented as mean ± SD, median and interquartile range or percentage as appropriate. Data adjusted for age, gender and diabetes duration were calculated by univariate linear regression. Comparisons between groups were performed using Student’s t test or one-way analysis of variance (ANOVA) followed by Bonferroni post hoc test, Mann–Whitney test, Kruskal–Wallis test or χ² test as appropriate. Missing values were not included for any calculations. All statistics were done in IBM SPSS Statistics for Windows, Version 24.0 (IBM Corp., Armonk, NY, USA).

Results

General and country-specific unadjusted demographic data are provided in Table 1, and Table 2 shows details on medical therapy not discussed in the article elsewhere. For further information, refer to the work by Eder et al.¹⁴

Table 1.

Unadjusted demographic data of the participants of the PROVALID study.

	Total	Austria	Hungary	Netherlands	Poland	Scotland
Age (years)	62.9 ± 10.0	64.3 ± 9.6	63.2 ± 9.6	65.0 ± 8.7	59.46 ± 10.9	60.23 ± 9.3
Body mass index (kg/m²)	30.1 (27.1; 33.9)	29.8 (26.9; 33.3)	30.5 (27.5; 34.3)	29.3 (26.4; 32.8)	29.4 (26.3; 32.8)	31.7 (28.1; 35.9)
Duration of hypertension (years)	12.0 (6.0; 18.0)	11.0 (5.0; 16.0)	13.0 (7.0; 20.0)	NA	11.0 (4.0; 17.0)	10.0 (7.0; 15.0)
Duration of diabetes (years)	8 (3; 13)	6 (3; 12)	12 (6; 18)	7 (4; 10)	3 (1; 10)	6 (3; 10)
Systolic/diastolic blood pressure (mm Hg)	137/80 ± 17/10	140/82 ± 19/10	134/78 ± 15/9	138/78 ± 17/10	139/84 ± 18/11	136/79 ± 17/10
Ex-smoker (%)	34.7	31.6	31.6	NA	39.5	42.1
Current smoker (%)	15.9	16.4	15.1	NA	16.9	16.0

Table 2.

Selected medical therapy of the study population (% of population).

	Total	Austria	Hungary	Netherlands	Poland	Scotland
Glucose-lowering agents
Biguanides	67.4	68.4	71.1	71.1	49.2	67.5
Insulins	30.3	12.5	44.0	12.0	23.6	8.2
Sulfonylureas	27.1	16.9	29.0	28.6	27.6	29.7
DPPIV inhibitors/GLP1 agonists	14.3	24.2	22.2	3.1	0.0	14.5
Blood pressure–lowering agents
Calcium antagonists	28.3	21.4	36.4	18.6	28.7	30.3
Alpha receptor blockers	9.7	2.1	19.7	0.0	11.2	5.7
Thiazide diuretics	35.0	30.8	47.2	30.5	29.2	19.4
Loop diuretics	11.4	6.1	19.5	4.1	11.4	8.4
Lipid-lowering agents
Statins	59.6	45.9	61.7	66.0	38.4	80.6

Prevalence of CKD and cardiovascular comorbidities

The percentage of patients in the different KDIGO stages of DKD (unadjusted as well as adjusted for age, gender and diabetes duration) in the total study population as well as the national sub-cohorts is given in Table 3.

Table 3.

Percentage of patients in the different KDIGO stages of DKD.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p value*
G1 or G2 and A1 (%)	65.2	64.8^a (60.3)	62.2^b (60.2)	64.8^c (77.5)	68.6^b (57.8)	68.5^b (71.6)	<0.001
G1 or G2 and A2 or A3 (%)	15.5	15.6^d (17.5)	15.1^b (12.8)	15.8^d (11.7)	15.3^a,c (23.1)	16.5^b (18.8)	<0.001
⩾Stage G3 and A1 (%)	12.8	13.1^a (15.3)	14.9^b (18.0)	12.9^c (9.1)	10.8^b (9.3)	9.6^b (6.0)	<0.001
⩾Stage G3 and A2 or A3 (%)	6.5	6.5^a (6.9)	7.8^b (9.1)	6.4^c (1.8)	5.3^b (9.7)	5.4^b (3.6)	<0.001
Any G and A1 (%)	78.0	77.9^a,e (75.6)	77.1^b (78.2)	77.8^c,e (86.6)	79.4^b (67.2)	78.1^c,d (77.6)	<0.001
Any G and A2 or A3 (%)	22.0	22.1^a,e (24.4)	22.9^b (21.8)	22.2^c,e (13.4)	20.6^b (32.8)	21.9^c,d (22.4)	<0.001
G1 or G2 (%)	80.7	80.4^a (77.8)	77.3^b (73.0)	80.6^c (89.1)	83.9^e (81.0)	85.0^d (90.3)	<0.001
⩾Stage G3 (%)	19.3	19.6^a (22.2)	22.7^b (27.0)	19.4^c (10.9)	16.1^b (19.0)	15.0^b (9.7)	<0.001

Significantly different from all countries except the Netherlands.

Significantly different from all countries.

Significantly different from all countries except Austria.

Significantly different from all countries except Poland.

Significantly different from all countries except Scotland.

Percentages are adjusted for age, gender and diabetes duration; unadjusted percentages are given in parentheses.

p values apply for adjusted as well as unadjusted data.

Even though the adjusted as well as unadjusted percentages of KDIGO stages differed statistically significantly between the countries, the populations were similar based on clinical judgement. About two-thirds of the subjects had normal renal function, 15% had maintained eGFR ≥60 mL/min/1.73 m² but at least moderately increased albuminuria (stages A2 and A3, former micro- and macroalbuminuria). In approximately 20%, eGFR was decreased <60 mL/min/1.73 m² (G3 or higher) and in this subgroup two-thirds were in albuminuria stage A1 and only 8% presented with a urinary albumin excretion >300 mg/g creatinine.

The percentage of patients with cardiovascular comorbidities in the total study population and the national sub-cohorts adjusted for age, gender and diabetes duration is presented in Table 4. The proportions in parentheses indicate the unadjusted cardiovascular comorbidity prevalence. The prevalence of comorbidities also differed significantly between the countries but in absolute terms these variations were small with some exceptions (e.g. less age, gender and diabetes duration adjusted coronary artery disease in the populations in Poland and Scotland). Supplementary Table 1 provides information on the prevalence of comorbidities in patients with normal renal function and those with CKD. Even though the latter were younger, their cardiovascular disease burden was significantly higher.

Table 4.

Percentage of patients with cardiovascular comorbidities.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p value*
Heart failure stage III/IV	2.5	2.4^a (4.3)	3.2^b (3.3)	2.4^c (0.0)	2.4^b (3.4)	1.6^b (1.0)	<0.001
Coronary artery disease	21.0	21.6^a,d (19.0)	22.7^c (21.5)	22.1^c (18.9)	15.9^b (27.2)	18.7^b (19.2)	<0.001
Peripheral artery disease	7.1	6.7^a,e (8.0)	8.8^b (8.9)	6.6^c (1.7)	5.4^b (9.1)	6.2^c (7.4)	<0.001
Cerebrovascular disease	6.7	6.8^a (10.1)	7.7 (7.9)	6.8^c (1.1)	5.3 (9.1)	5.8 (5.7)	<0.001

Significantly different from all countries except the Netherlands.

Significantly different from all countries.

Significantly different from all countries except Austria.

Significantly different from all countries except Hungary.

Significantly different from all countries except Scotland.

Percentages are adjusted for age, gender and diabetes duration; unadjusted percentages are given in parentheses.

p values apply for adjusted as well as unadjusted data.

Metabolic control

The KDIGO guidelines recommend a target HbA1c value of 7% to prevent or delay progression of microvascular complications of diabetes including DKD. In individuals with comorbidities or limited life expectancy, the goal should be extended to >7%, whereas in those at risk for hypoglycaemia, HbA1c levels should not be below 7%¹⁰. The ADA¹¹ recommends an HbA1c level around 7% for all T2DM patients except pregnant women but with individualization according to a number of factors. In those with a short disease duration, treatment with lifestyle measures or metformin only, a long life expectancy or no significant cardiovascular comorbidity the goal can be set for <6.5%. On the contrary, those with a risk for hypoglycaemia, advanced micro- or macrovascular disease-extensive comorbidities or long-standing diabetes might aim for <8%. Table 5 shows data on the median HbA1c levels and Table 6 shows the percentage of patients within specific ranges in the total study population and the national sub-cohorts. In all countries except Scotland, more than 80% of subjects included had an HbA1c of <8% suggesting that in general metabolic control was acceptable in the majority of patients.

Table 5.

Median HbA1c levels and the percentage of patients within specific ranges.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p
HbA1c	6.8 (6.2; 7.5)	6.7 (6.2; 7.4)^a,b	6.9 (6.2; 7.7)^a,c	6.7 (6.4; 7.3)^b,c	6.5 (5.9; 7.2)^d	7 (6.4; 8)^d	<0.001

Significantly different from all countries except the Netherlands.

Significantly different from all countries except Hungary.

Significantly different from all countries except Austria.

Significantly different from all countries.

Percentages are adjusted for age, gender and diabetes duration; unadjusted percentages are given in parentheses.

Table 6.

Percentage of patients within specific ranges of HbA1c levels. Analysis done be Chi-square test, no poshoc analysis available.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p
<6 (%)	15.6	13.4	18.0	7.5	27.0	13.3	<0.001
6 to <8 (%)	69.6	74.5	63.4	84.6	61.6	62.7	<0.001
8 to <10 (%)	12.3	10.4	15.6	7.4	9.7	17.2	<0.001
>10 (%)	2.5	1.7	3.0	0.5	1.7	6.8	<0.001

Percentages are adjusted for age, gender and diabetes duration; unadjusted percentages are given in parentheses.

Blood pressure control

Both KDIGO¹⁰ and the ADA guidelines¹² recommend that blood pressure in patients with T2DM and urinary albumin excretion below 30 mg/day should not exceed 140/90 mm Hg. In subjects with an elevated urinary albumin excretion (A2 and A3), the goal blood pressure is reduced to 130/80. Table 7 shows the mean blood pressure values for the various groups and Table 8 shows the percentage of subjects within these recommendations in the total study population and in the various national sub-cohorts. Whereas about 75% of subjects reached goal blood pressure in diabetic patients with normal or mildly increased albuminuria (A1), blood pressure was <130/80 mm Hg in only 10%–20% of patients with higher urinary albumin excretion. Interestingly these patients not only reached goal blood pressure values less often but even had higher mean systolic blood pressure in all countries expect Poland when compared to normoalbuminuric study participants. Worse blood pressure control was achieved in patients in stages A2 and A3 when compared to A1 despite the fact that antihypertensive therapy was intensified (see Supplementary Table 2). Further analysis revealed that worse blood pressure control was observed in all countries expect Poland in patients at early stages (G1 and G2) of renal disease and at least moderately increased albuminuria (A2 and A3) (Supplementary Table 3) but not in subjects with more advanced renal failure.

Table 7.

Mean blood pressure values.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p value
Systolic/diastolic blood pressure (mm Hg)	137/80 ± 17/10	140^a/82^a,b,c ± 19/10	134^d/78^e,f ± 15/9	138^b/78^a,e,f ± 17/10	139^b/84^b,c ± 18/11	136^a/79^e,f ± 17/10	<0.001
Systolic/diastolic blood pressure (mm Hg) in subjects with urinary albumin excretion <30mg/g creatinine	136/79 ± 17/10	139^a/81^g ± 18/9	134^h/78^f ± 15/9	138^a/78^e,f ± 17/10	138^a/83^b,c ± 18/11	135/78^e,f ± 16/10	<0.001
Systolic/diastolic blood pressure (mm Hg) in subjects with urinary albumin excretion ⩾30 mg/g creatinine	140/81 ± 18/11	145^a/84^a,b,c ± 20/10	136^f/78^e,f ± 17/10	141/79^e,f ± 20/11	139/84^b,c ± 17/12	141/80^e,f ± 19/13	<0.001
p values for blood pressure urinary albuminuria <30 versus ⩾30 mg/g creatinine and for systolic–diastolic blood pressure	<0.001/0.006	0.001/0.024	0.009/0.401	0.042/0.477	0.363/0.796	0.005/0.155

Significantly different from Hungary.

Significantly different from Scotland.

Significantly different from the Netherlands.

Significantly different from all countries.

Significantly different from Poland.

Significantly different from Austria.

Significantly different from all countries except Poland.

Significantly different from all countries except Scotland.

Table 8.

Percentage of subjects reaching recommended blood pressure goal levels.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p value
Albuminuria <30 mg/g creatinine and blood pressure <140/90 mm Hg (%)	74.9	71.9	81.8	67.9	75.9	70.2	<0.001
Albuminuria ≥30 mg/g creatinine and blood pressure <130/80 mm Hg (%)	14.9	15.4	17.5	8.3	19.8	12.6	<0.001

Prescription of agents blocking the renin–angiotensin–aldosterone system, statins and aspirin

Both KDIGO¹⁰ and the ADA¹² recommend ACE inhibitor or angiotensin receptor blocker (ARB) therapy in patients with T2DM and albuminuria >30 mg/g creatinine. Table 9 shows the percentage of subjects treated with these agents in the total study population and in the national subpopulations according to their albuminuria status.

Table 9.

Use of agents that block the renin–angiotensin system.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p value
% of patients treated with ACE inhibitors/ARBs	76.0	72.1	87.4	66.7	69.5	72.8	<0.001
% of patients in any G but A1 stage treated with ACE inhibitors/ARBs	74.1	69.1	86.9	65.2	67.2	68.6	<0.001
% of patients in any G stage but A2 or A3 stage treated with ACE inhibitors/ARBs	84.6	81.3	94.9	75.7	73.9	88.3	<0.001
p values for any G and A1 versus any G and A2 or A3	<0.001	0.004	<0.001	0.027	0.117	<0.001

About three-quarters of all subjects included received RAS blocking therapy (66.7%–87.4%). The use of the agents was significantly higher in patients with KDIGO stages A2 and A3 in all countries expect in Poland. In addition, patients with KDIGO stages G3 or higher were also preferentially treated by RAS blockade (with the exception of expect for Poland and The Netherlands; Supplementary Table 4a). A more detailed analysis of RAS blocking therapy according to different KDIGO classification groups is provided in Supplementary Table 4b. RAS blocking therapy was more prevalent in subjects in stages A2 or A3 in G 1 and 2 stages of the disease. The same trend could not be observed in more advanced G stages, if anything the frequency of administration was lower in the presence of moderately or severely increased albuminuria. Additional analysis of this subgroup showed that normoalbuminuric patients with eGFR values below 60 mL/min/1.73 m² had higher mean eGFR but lower blood pressure than their albuminuric counterparts [50.8 mL/min/1.73 m², interquartile range (IQR) 42.4–55.2 vs 44.0 mL/min/1.73 m², IQR 32.8–57.4; p < 0.001; 136/77 ± 17/10 vs 138/79 ± 18/11; p = 0.09/0.01]. In addition, the serum potassium levels were identical (4.5 mmol/L, IQR 4.2–4.8 vs 4.5 mmol/L, IQR 4.1–5.0; p = 0.277). Thus, based on these cross-sectional data, we cannot explain the lower use of ACE/ARB therapy in patients with advanced renal disease and moderately or severely increased albuminuria. It could be speculated that some initially albuminuric subjects had regressed to normoalbuminuria but stayed on RAS blocking therapy.

As far as cholesterol-lowering therapy is concerned, the ADA guidelines recommend that all diabetic patients over the age of 40 should receive a statin¹². KDIGO suggests statin therapy for all patients with CKD¹⁰. Table 10 and Supplementary Table 5 show details on adherence to these guidelines. In Table 11, the use of aspirin is given, which is recommended for all diabetic patients >50 years of age by the ADA¹³ and for all subjects with CKD by KDIGO.¹⁰ Supplementary Table 6 shows more details on the use of ASA according to guidelines and on serum cholesterol levels, which were significantly different between the countries. A total of 3902 subjects were older than 40 years and thus generally eligible for statin therapy according to the ADA guidelines. In countries with high use of statins, the presence of renal disease did not affect statin prescription frequency. However, in the two countries with less than 50 statin use in general (Austria and Poland), subjects with renal disease were more likely to be prescribed statin therapy.

Table 10.

Serum cholesterol levels and use of statins according to guidelines and in specific subpopulations.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p value (countries)
Serum total cholesterol (mg/dL)	176 (147; 208)	194^a (166; 224)	182^b (151; 216)	162^c (137;185)	183^d (150; 224)	166^e (143; 189)	<0.001
% of subjects >40 years of age treated with statins	60.5	46.3	62.8	66.1	39.8	82.7	<0.001
% of subjects in stage G1 or G2 and A1 treated with statins	58.9	40.5	61.6	66.7	33.5	79.7	<0.001
% of all other subjects treated with statins	60.6	53.1	62.5	66.3	44.7	84.4	<0.001
p values for G1 or G2 and A1 vs all others	0.297	0.002	0.741	0.929	0.009	0.230

Significantly different from all countries except Poland.

Significantly different from all countries.

Significantly different from all countries except Scotland.

Significantly different from all countries except Austria.

Significantly different from all countries except the Netherlands.

Table 11.

Use of ASA according to guidelines and in specific subpopulations.

	Total	Austria	Hungary	Netherlands	Poland	Scotland	p value (countries)
% of subjects >50 years of age treated with ASA	28.9	26.8	35.5	5.2	43.8	43.6	<0.001
% of subjects in stage G1 or G2 and A1 treated with ASA	23.2	21.6	29.6	4.2	35.8	34.6	<0.001
% of subjects all other subjects treated with ASA	34.9	29.4	39.7	7.8	41.6	52.5	<0.001
p value for G1 or G2 and A1 vs all others	<0.001	0.028	<0.001	0.047	0.173	<0.001

In general, patients on statin therapy were older, more likely to be male and had significantly lower cholesterol levels than those not treated. Their cardiovascular comorbidity was much higher leading to the conclusion that prescription of statin therapy was driven rather by cardiovascular comorbidity than guideline recommendation.

Overall, ASA use was low in all countries, of the 3560 patients older than 50 years only about 29% were treated. However, a huge difference between the countries was observed. On average, patients with renal disease were more likely to be treated. Similar to statin therapy, the administration of ASA was mostly driven by the presence of cardiovascular disease and especially coronary artery, peripheral or cerebrovascular disease. Interestingly, patients with advanced heart failure were significantly less likely to receive ASA-containing drugs.

Discussion

In general, our study shows that there is room for improvement regarding guideline adherence at the primary level of healthcare in patients with T2DM at least in the European countries included in PROVALID. Blood pressure control was poor in patients with elevated albuminuria with only 15% of all subjects reaching the recommended target values of <130/80 mm Hg. Both ADA and KDIGO suggest statin therapy in patients with T2DM or CKD without taking care of serum cholesterol levels, yet only about 60% of participants received medication. The rate of prescription was even lower for ASA (approximately 30%). Our data suggest that contrary to guideline recommendations, the latter two treatments were rather used because of the presence of cardiovascular comorbidities than in a ‘primary prevention’ effort.

An additional interesting aspect is the large variability of therapeutic practice pattern between the countries despite the fact that the populations recruited were quite similar with regard to patient history and the prevalence of risk factors and comorbidities. One reason for the observed diversity could be that GPs rather follow national guidelines. However, the Austrian Society of Nephrology has endorsed the KDIGO guidelines and the Austrian Diabetes Association¹⁵ issued a statement that also closely follows the ADA/KDIGO suggestions. The situation is similar in Scotland,¹⁶ Poland¹⁷ and Hungary.^18,19 Guidelines in the Netherlands also mostly follow KDIGO and ADA but suggest even lower blood pressure goals. However, statin therapy is recommended only in patients with serum cholesterol levels >2.5 mmol/L and thiazide diuretics should be administered as first-line therapy in patients without micro- or macroalbuminuria.²⁰ Neither the Netherlands nor Scotland local guidelines specifically recommend aspirin therapy for patients with T2DM.

Available evidence-based guidelines provide targets and performance measures for the treatment of T2DM patients, but obviously, a wide gap separates guidelines-driven recommendations from their clinical application, a phenomenon hindering the transfer of proven benefits to affected populations. In this regard, the results of our study replicate the findings of other authors in patients with diabetes and comorbidities like coronary artery or cerebrovascular disease. Bohn et al.²¹ studied almost 30,000 German and Austrian patients with T2DM with respect to secondary prevention after myocardial infarction (MI) and stroke. HbA1c goal according to national guidelines <7.5% was achieved in 64.9% (MI) and in 61.1% (stroke) of patients; LDL-C <100 mg/dL was documented in 56.2% (MI) and in 42.2% (stroke). Target values of blood pressure (<130/80 mm Hg in MI, 120/70–140/90 in stroke) were reached in 67.0% (MI) and in 89.9 % (stroke). Prescription prevalence of inhibitors of platelet aggregation was 50.7% (MI) and 31.7% (stroke). About 57% (MI) and 40.1% (stroke) used statins. Similar data were reported by Modesti et al.²² in Italy and Renard et al.²³ in Luxembourg. A total of 6187 patients with coronary artery disease were investigated in EUROASPIRE IV in 24 European countries 2012–2013. About 46% had no diabetes, 19% were newly diagnosed at enrolment and 35% were known diabetics. The optimum use of cardio-protective drugs in these groups was 53%, 55% and 60%, respectively. A blood pressure target of <140/90 mm Hg was achieved in 68%, 61% and 54%, and an LDL-C target of <1.8 mmol/L was achieved in 16%, 18% and 28%. Despite the fact that compared with data from previous surveys the use of cardio-protective drugs had increased and more patients were achieving the risk factor treatment targets, the authors concluded that there is further potential to improve the management of patients with diabetes and coronary artery disease.²⁴

Several efforts have been tested for their capacity to improve guideline adherence. In the United States, Kuo et al. compared care provided to patients with diabetes mellitus by nurse practitioners (NPs) and primary care physicians (PCPs) and found no significant difference. NPs and PCPs had similar rates of LDL-C testing and nephropathy monitoring, but NPs had lower rates of eye examinations and HbA1c testing. NPs were more likely to have consulted cardiologists, endocrinologists, and nephrologists. There was no statistically significant difference in adjusted Medicare spending between the two groups.²⁵

Another approach taken is to use incentive-based primary care, a fee for service-funding model. In British Columbia, these incentives, after controlling for patients’ age, sex, service needs level and continuity of care (defined as attachment to a general practice), reduced the net annual healthcare costs for patients with hypertension, chronic obstructive pulmonary disease and congestive heart failure, but not diabetes mellitus.²⁶ A recent Cochrane Database review concluded that the use of financial incentives is growing but the authors put forward that there is insufficient evidence to support or not support this trend. Implementation should proceed with caution and incentive schemes should be more carefully designed before implementation.²⁷ In line with this Boyd et al.²⁸ concluded for a group of elderly multi-morbid subjects pay for service incentives and the resulting adherence to guidelines lead to undesired polypharmacy and inadequate care. In this context, it is interesting to note that during the period of recruitment into PROVALID, primary care providers received financial reimbursement in Scotland for meeting audit and/or treatment targets for patients with diabetes under the arrangements of the Quality and Outcomes Framework (QOF). QOF is intended to standardize improvements in the delivery of primary medical services. For diabetes, specific QOF targets for reimbursement during the period of study included percentage of diabetic patients registered with that primary care provider with blood pressure <150/90 mmg and/or <140/80 mm Hg, respectively, percentage treated with an ACE inhibitor (or ARB) in the presence of micro/macroalbuminuria, percentage with IFCC HBA1c at various thresholds (59, 64 and 75 mmol/mol) and percentage with cholesterol <5 mmol/L.

Electronic health records might be an interesting tool to improve guideline adherence. In a study by Ancker et al.²⁹ (183,095 visits with 61,977 patients), condition-specific best-practice alerts and order sets were associated with better scores on clinical quality measures capturing processes in diabetes, cancer screening, tobacco cessation, and pneumonia vaccination. For example, providers above the median in use of tobacco-related alerts had higher performance on tobacco cessation intervention metrics and providers above the median in use of diabetes order sets had higher quality on diabetes LDL testing. However, post hoc examination of the results showed that the positive associations were with measures of healthcare processes (such as rates of LDL testing), whereas there were no positive associations with measures of healthcare outcomes (such as LDL levels).²⁹ Similar effects were reported by Benkert et al.³⁰

Finally, CME efforts might be applied. About 53% of 383 physicians completing a questionnaire reported using diabetes guidelines routinely. Mean diabetes knowledge score for guideline users was significantly higher than non-guideline users. The former were significantly more likely to report a good understanding of type 2 diabetes medications, less likely to report their unfamiliarity with insulin as an important barrier to early insulin use and with pharmacologic options as a barrier to prescribing intensive multifactorial interventions. Associations remained significant after adjusting for physician specialty, practice volume and frequency diagnosing or treating diabetes patients.³¹

Conclusion

In summary, our study shows that international society guidelines for patients with T2DM are only partially followed at the primary level of healthcare in the European countries included in our study. In addition, treatment applied varies widely. It cannot be answered by our cross-sectional analysis if this diversity affects clinical outcomes. Only longitudinal follow-up data from our cohort will help us to understand whether our findings contribute to the reported national variability in the incidence of ESRD.

Supplemental Material

dvdres-feb-2018-00028-File003 – Supplemental material for Guidelines and clinical practice at the primary level of healthcare in patients with type 2 diabetes mellitus with and without kidney disease in five European countries

Supplemental material, dvdres-feb-2018-00028-File003 for Guidelines and clinical practice at the primary level of healthcare in patients with type 2 diabetes mellitus with and without kidney disease in five European countries by Susanne Eder, Johannes Leierer, Julia Kerschbaum, Laszlo Rosivall, Andrzej Wiecek, Dick de Zeeuw, Patrick B Mark, Georg Heinze, Peter Rossing, Hiddo L Heerspink and Gert Mayer in Diabetes & Vascular Disease Research

Footnotes

Acknowledgements

S.E. and J.L. have contributed equally.

Author’s contribution

All authors contributed to writing of the paper, results and discussion. S.E, J.L, J.K., G.H. and G.M. conducted data analysis. All authors except J.L. and J.K. were responsible for study design and conduct. H.L.H, L.R., A.W., P.B.M. and G.M. are national coordinators.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was in part funded by the European Union (7th Framework grant number 241544: ‘Systems Biology towards Novel Chronic Kidney Disease Diagnosis and Treatment’) and AbbVie.

Availability of data

The data sets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Ethical approval

The PROVALID study protocol was approved by the local IRB in each participating country.

ORCID iDs

Johannes Leierer

Gert Mayer

Supplemental material

Supplemental material for this article is available online.

References

GBD 2013 mortality and causes of death collaborators. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2015; 385: 117–171.

Kainz

Hronsky

Stel

et al . Prediction of prevalence of chronic kidney disease in diabetic patients in countries of the European Union up to 2025. Nephrol Dial Transpl 2015; 30: iv113–iv118.

Jha

Garcia-Garcia

Iseki

et al . Chronic kidney disease: global dimension and perspectives. Lancet 2013; 382: 260–272.

Vanholder

Lameire

Annemans

et al . Cost of renal replacement: how to help as many as possible while keeping expenses reasonable? Nephrol Dial Transpl 2016; 31: 1251–1261.

Kramer

Pippias

Stel

et al . Renal replacement therapy in Europe: a summary of the 2013 ERA-EDTA Registry annual report with a focus on diabetes mellitus. Clin Kidney J 2016; 9: 457–469.

Brück

Stel

Gambaro

et al . CKD prevalence varies across the European general population. J Am Soc Nephrol 2016; 27: 2135–2147.

Tanner

Gutiérrez

Judd

et al . Geographic variation in CKD prevalence and ESRD Incidence in the United States: results from the reasons for geographic and racial differences in stroke (REGARDS) study. Am J Kidney Dis 2013; 61: 395–403.

Zhang

Wang

et al . Prevalence of chronic kidney disease in China: a cross-sectional survey. Lancet 2012; 379: 815–822.

Brück

Jager

Dounousi

et al . Methodology used in studies reporting chronic kidney disease prevalence: a systematic literature review. Nephrol Dial Transplant 2015; 30: iv6–iv16.

10.

KDIGO. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013; 3: 1.

11.

Association AD. 5. Glycemic targets. Diabetes Care 2016; 39: S39–S46.

12.

Association AD. 8. Cardiovascular disease and risk management. Diabetes Care 2016; 39: S60–S71.

13.

Association AD. 9. Microvascular complications and foot care. Diabetes Care 2016; 39: S72–S80.

14.

Eder

Leierer

Kerschbaum

et al . A prospective cohort study in patients with type 2 diabetes mellitus for validation of biomarkers (PROVALID): study design and baseline characteristics. Kidney Blood Press R 2018; 43: 181–190.

15.

Österreichische Diabetes Gesellschaft. http://www.oedg.at/ (2017, accessed 12 September 2017).

16.

SIGN 116 Management of Diabetes, https://www.sign.ac.uk/sign-116-and-154-diabetes.html (accessed 24 August 2018).

17.

Diabetology

2018 Guidelines on the management of diabetic patients. A position of Diabetes Poland. Clin Diabetol 2018; 7: 1–90.

18.

Gaal

Gerö

Hidvégi

et al . Egészségügyi szakmai irányelv: a diabetes mellitus kórismézéséről, a cukorbetegek antihyperglykaemiás kezeléséről és gondozásáról felnőttkorban. Diabetol Hung 2017; 25: 3–77.

19.

Magyar Nephrologiai Társaság, http://www.nephrologia.hu/info.aspx?sp=17 (accessed 25 September 2017).

20.

De Grauw

Kaasjager

Bilo

et al . Landelijke Transmurale Afspraak Chronische nierschade. Huisarts En Wet 2009; 52: 586–595.

21.

Bohn

Schöfl

Zimmer

et al . Achievement of treatment goals for secondary prevention of myocardial infarction or stroke in 29,325 patients with type 2 diabetes: a German/Austrian DPV-multicenter analysis. Cardiovasc Diabetol 2016; 15: 72.

22.

Modesti

Bartaloni

Bellagamba

et al . Health care delivery in type 2 diabetes. A survey in an Italian primary care practice. Prim Care Diabetes 2015; 9: 9–14.

23.

Renard

Bocquet

Vidal-Trecan

et al . Adherence to international follow-up guidelines in type 2 diabetes: a longitudinal cohort study in Luxembourg. PLoS ONE 2013; 8: e80162.

24.

Gyberg

De Bacquer

De Backer

et al . Patients with coronary artery disease and diabetes need improved management: a report from the EUROASPIRE IV survey: a registry from the Euroobservational Research Programme of the European Society of Cardiology. Cardiovasc Diabetol 2015; 14: 133.

25.

Kuo

Y-F

Goodwin

Chen

N-W

et al . Diabetes mellitus care provided by nurse practitioners vs primary care physicians. J Am Geriatr Soc 2015; 63: 1980–1988.

26.

Hollander

Kadlec

Incentive-based primary care: cost and utilization analysis. Perm J 2015; 19: 46–56.

27.

Scott

Sivey

Ait Ouakrim

et al . The effect of financial incentives on the quality of health care provided by primary care physicians. In: Higgins

Green

(eds) Cochrane database of systematic reviews. Hoboken, NJ: John Wiley & Sons, Ltd., 2011.

28.

Boyd

Darer

Boult

et al . Clinical practice guidelines and quality of care for older patients with multiple comorbid diseases: implications for pay for performance. JAMA 2005; 294: 716–724.

29.

Ancker

Kern

Edwards

et al . Associations between healthcare quality and use of electronic health record functions in ambulatory care. J Am Med Inform Assoc 2015; 22: 864–871.

30.

Benkert

Dennehy

White

et al . Diabetes and hypertension quality measurement in four safety-net sites: lessons learned after implementation of the same commercial electronic health record. Appl Clin Inform 2014; 5: 757–772.

31.

Corriere

Minang

Sisson

et al . The use of clinical guidelines highlights ongoing educational gaps in physicians’ knowledge and decision making related to diabetes. BMC Med Educ 2014; 14: 186.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB