Level of glycaemic and lipid control among diabetic patients in Greek primary care

Introduction

Achieving glycaemic and lipid goals in diabetic patients has always been a problem, especially in primary care [1]. In Greece, diabetes mellitus (DM) is common [2]; however, there are no reported data on the control of DM in primary care. Our aim was to evaluate the level of glycaemic and lipid control among diabetic patients visiting a primary care setting.

Methods

A random sample of 302 patients with DM (98% with type 2) was recruited between January 2003 and May 2005; they constitute about 15% of the estimated population with diagnosed diabetes in the area covered by our health centre [2]. In all patients, weight, height, and waist and hip circumferences were measured, and blood was taken, after overnight fast, for the measurement of glycosylated haemoglobin (HbA_1c) and lipids. Characteristics of the patients included a mean age of 65±10 years, a mean disease duration of 8±7 years, a mean BMI of 31±5 kg/m² and a mean waist circumference of 105±12 cm. Student's t-test, one-way analysis of variance and the Pearson correlation test were used for statistical analyses.

Results

The mean HbA_1c in the whole population was 8.1±1.8%. Glycaemic control was optimal in 30.5% (HbA_1c<7%), fair in 23.2% (HbA_1c 7–7.9%) and poor in 46.3% (HbA_1c≥8%) of patients. For lipid levels, only 53 (17.5%) were at recommended levels (<100 mg/dl) for low-density lipoprotein cholesterol (LDL-C); when the cut-off point was designated to be 130 mg/dl, this proportion increased to 37%. About 26% (n=79) had total cholesterol (TC) < 200 mg/dl, 171 (56.6%) had high-density lipoprotein cholesterol (HDL-C) > 45 mg/dl, and 176 (58.3%) had triglyceride values <150 mg/dl. Among those who were “off target” (LDL-C ≥ 100 mg/dl), only 25% were treated.

Better glycaemic control was related to shorter duration of DM (r=0.26, p<0.0001) and lower value of waist-to-hip ratio (r=0.12, p=0.035), as expected. Moreover, as shown in Table I, the mean value of HbA_1c was lower in patients managed with diet alone in comparison to those taking oral agents and/or insulin. With regard to lipid control, only HDL-C was found to be significantly lower in men (p=0.0001) and among those with confirmed coronary artery disease (CAD) (p=0.035). Here, it is of particular interest to note that, although the percentage of patients with CAD who were treated with statin was higher in comparison to those without CAD (30% vs 16%, p=0.03), the proportion of those reaching the recommended level of LDL-C was almost the same (17% vs 17.5%). Age, BMI, occupational status, family history of diabetes and smoking were not significantly related to glycaemic or lipid level.

Table I.  Differences in HbA_1c and lipid levels (mean values).

Variable	HbA1c	TC	LDL-C	HDL-C	Triglycerides
Age (y)
31–65 (n=143)	8.1	229.7	140.1	46.6	187.1
66–86 (n=159)	8.2	220.5	138.9	48.5	149.7
Gender
Female (n=165)	8.2	229.1	142.1	51.4^a	171.3
Male (n=137)	8.0	217.2	130.6	43.1	160.8
Therapy
Diet alone (n=52)	7.1 ^a	236.5	150.6	49.9	149.1
Oral agents (n=202)	8.2	221.1	137.6	46.8	177.4
Insulin (n=48)	9.1	226.4	134.9	48.3	140.4
BMI (kg/m^2)
< 25 (n=35)	8.1	206.2	125.5	47.5	116.1
25–29.9 (n=104)	8.4	228.1	145.3	48.2	176.2
≥ 30 (n=163)	8.0	226.7	141.6	47.6	174.5
Coronary artery disease
No (n=242)	8.1	225.9	140.7	48.5	166.9
Yes (n=60)	8.3	218.4	134.1	43.2	166.3
Hypertension
No (n=84)	7.9	225.4	139.3	49.2	148.4
Yes (n=218)	8.3	224.4	139.6	46.9	174.2

^ap<0.001; ^bp=0.035.

Discussion

These findings indicate that the glycaemic and lipid control of Greek diabetic patients in primary care is largely suboptimal. Given the study sample was randomly selected from a part of the Attica region, in which there lives about 35% of the Greek general population, we believe that our results could be generalized for the entire DM population of Greece. However, referral at only one primary care centre is an important limitation of this study.

Specific patient characteristics seem to play a limited role in glycaemic and especially lipid control. Hence, inadequate metabolic control is mainly due to a deficiency of the Greek primary care system to provide optimized care for diabetic patients. Compared to other European countries, metabolic control in Greek diabetic patients appears to be of a lower level compared to northern European countries with highly developed general practice systems [1], [3]. In light of the relatively scarce data from other regions of Europe, it seems that our findings are similar to those reported from other southern European countries such as Spain and Cyprus [4], [5]. Interestingly, these countries, like Greece, are characterized by a less developed primary healthcare system and high prevalence rates of DM.

These observations should alert primary care providers to develop better management strategies and methods to improve the quality of diabetes care, especially in European countries with a high prevalence of diabetes. More specifically, efforts should focus on improving diabetes registration, on appropriate referrals to specialized professionals and on continuous intervention for lifestyle changes of diabetic patients [1], [3], [5]. Also, this study suggests the need for further research on the specific factors influencing the quality of diabetes care in the primary care setting.