Introduction

The traditional Mediterranean diet (MD) has emerged as a model of healthy eating on the basis of a large body of epidemiological studies performed during the last 50 years [1,2,3,4,5]. Recent intervention studies have also proved that the MD has favourable effects on plasma lipoprotein levels, endothelial function, insulin sensitivity, incidence of type 2 diabetes, cardiovascular events and cancer [6,7,8]. However, data on the impact of a dietary intervention based on Mediterranean tradition on weight reduction are still scant [9, 10]. Generally, dietary strategies for the correction of overweight/obesity are not very effective since, in most cases, people undertaking weight-reducing regimens experience a transient weight reduction followed by a progressive weight regain after few months [11,12,13].

Therefore, the aim of this study was to evaluate the effectiveness of an intensive dietary treatment aimed at implementing the traditional MD in a group of individuals with high cardiometabolic risk.

The major innovative feature of our intervention consisted of a persistent adherence to the traditional MD, achieved by providing participants with a 3-month supply of meals of the traditional MD, prepared under the supervision of a nutritionist by a local restaurant where they consumed lunch daily. The efficacy of the intervention was evaluated by measuring the change in body weight and its impact on cardiometabolic parameters up to 9 months after the end of the intervention. The programme was part of a healthy education project aiming at controlling energy intake and promoting food choices typical of the traditional MD.

Materials and methods

Subjects

As part of the project, 116 subjects (64 men and 52 women) were recruited by the general practitioners (GPs) of the city of Salerno on the basis of the inclusion and exclusion criteria listed in Table 1. Those who fulfilled the inclusion criteria and did not present any of the exclusion criteria were enrolled and invited to the GP clinic or to the Operative Territorial Unit for the care of diabetes, where they received detailed information on the procedures of the study and signed the informed consent to participation.

Table 1 Selection criteria for the study participants
Full size table

Study design

The study was designed as a non-randomized study with a 3-month intervention and a 9-month follow-up. Participants were assigned to the intensive or conventional treatment on the basis of their commitment to consume at least one meal per day at the restaurant affiliated with the study project; people with the same clinical and metabolic features but unable to comply with this requirement for personal or family reasons were assigned to the control group. Special attention was paid to ensure that the two groups were comparable with regard to clinical characteristics and cardiovascular risk factor profile. Of the 116 subjects recruited for the study, 71 were assigned to the intensive intervention and 45 to the conventional intervention. All participants underwent a detailed clinical examination (assessment of weight, height, waist circumference, blood pressure by standard methods) and a fasting blood sampling for measurement of plasma glucose, triglycerides, total and HDL cholesterol, glycated haemoglobin (HbA1c). These evaluations were repeated at the end of intervention and at 3- and 9-month follow-up. In addition, information on the participants’ dietary habits was collected through the compilation of a 7-day food record at baseline, at the end of the intervention and during the follow-up. Dietary energy intake and nutrient composition were calculated using a software containing Italian food composition tables [14].

Intensive intervention

The intensive intervention was based on two pillars: (1) an educational programme to healthy lifestyle (eating behaviour) based on twelve weekly group meetings held by a nutritionist and (2) free-of-charge daily supply of two main meals prepared for the purpose of the study. One meal was served at the Restaurant “La mia tavola”, and the other one was delivered to participants’ home. The meals were prepared, under the supervision of an expert nutritionist, by a chef trained on the principles of the Mediterranean diet tradition. All menus were based on food produced locally. The diet composition complied with the recommendations for the treatment and prevention of diabetes and cardiovascular diseases: energy (1500–1700 kcal), vegetable fibre (15–20 g/1000 kcal), total carbohydrates (45–55%), total mono-disaccharides (< 15%), total fat (30–35%), saturated fat (< 8%), monounsaturated fat (15–20%), protein (15–20%), sodium (< 1500 mg) [15]. Once a week, participants joined the chef to assist in the preparation of the meals and learn the recipes and the cooking methods of the traditional Mediterranean cuisine of Southern Italy.

Conventional intervention

The conventional intervention consisted of: (1) an individual education session with the nutritionist who provided each participant a grid of recommended food choices and (2) reinforcements of nutritional messages by the GP on a monthly basis during the 3-month intervention. Dietary prescription to the control group was based on the same nutritional recommendations followed for the meal preparation of the intervention group [14].

All participants gave their informed consent to participate, and the study protocol was approved by the Ethics Committee of the “Federico II” University of Naples.

Analyses of samples

Plasma glucose, total cholesterol, high-density lipoprotein (HDL) cholesterol and triglyceride concentrations were assayed according to standard procedures. LDL cholesterol was calculated using Friedewald’s formula. Glycated haemoglobin was measured by high-performance liquid chromatography (HPLC). All analyses were performed at a Central Laboratory.

Definitions

Diabetes was defined by the American Diabetes Association diagnostic criteria as fasting blood glucose ≥ 126 mg/dl (7.0 mmol/L) and/or use of hypoglycaemic drugs. Hypertension was defined as blood pressure ≥ 140 and/or 90 mmHg or use of antihypertensive treatment. Hypercholesterolaemia was defined as fasting LDL cholesterol ≥ 130 mg/dl (3.36 mmol/L) or use of hypolipidemic drugs.

Statistical analysis

All data are expressed as mean ± standard deviation (M ± SD). Differences with respect to baseline variables and to the changes during the study between the two groups were assessed by t tests for independent samples. Changes in BMI and other clinical and biochemical variables within each group were tested using paired Student’s t test corrected for multiple comparisons. A p value < 0.05 was considered statistically significant. Statistical analysis was performed with the SPSS 17 system (SPSS Inc).

Results

The main characteristics of the intervention and the control groups at baseline are reported in Table 2. The whole population was characterized by an excess body weight as indicated by an average BMI of 32 kg/m2 in both groups. Fasting plasma glucose was above normal in both groups indicating the presence of impaired glucose regulation in most participants. No difference was found between the two groups with regard to the other variables. Twenty-four (34%) subjects of the intensive intervention group and 16 (35%) subjects of the control group had a previous diagnosis of type 2 diabetes. In the intervention group, twenty-two were treated with metformin alone or in combination with sulphonylureas and/or DPP-4 inhibitors and two were treated with diet alone. In the control group, thirteen participants received metformin alone or in combination with sulphonylureas and/or DPP-4 inhibitors and three were on diet alone. Hypertension was present in 74 and 52% of subjects in the intervention and the control group, respectively. Hypercholesterolaemia was present in 46% of the intervention group and 43% of the control group.

Table 2 Clinical characteristics of participants at baseline
Full size table

Effects of the intervention

At the end of the intervention, 10 out of 45 control subjects abandoned the study. All participants in the intervention group completed the educational and experiential programme.

Compared to pre-intervention, control subjects showed a significant reduction in body weight, waist circumference and arterial blood pressure (p < 0.05–p < 0.001) (Table 3). The intensive treatment group showed a significant reduction in body weight, waist circumference, glycated haemoglobin, fasting plasma triglycerides, arterial blood pressure and a significant increase in HDL cholesterol (Table 3). Notably, the changes (Δ) of the clinical and metabolic parameters after the intervention were of much greater extent and clinical impact in the intensive intervention compared with the control group. Indeed, the changes were as follows: BMI (− 2.50 vs. − 0.3 kg/m2, p < 0.001), waist circumference (− 9.5 vs. − 1.6 cm; p < 0. 001), plasma glucose (− 0.3 vs. + 0.5 mmol/L; p < 0.002), triglycerides (− 0.3 vs. + 0.2 mmol/L; p < 0.002), systolic blood pressure (− 13 vs. − 7 mmHg; p < 0.01) and diastolic blood pressure (− 9 vs. − 4 mmHg; p < 0.001) (Table 3). The improvement in glucose metabolism was particularly evident in participants with type 2 diabetes; in fact, HbA1c level decreased significantly in the intensive treatment group (from 61 ± 5 to 51 ± 6 mmol/mol; p < 0.0001), while it remained substantially unchanged in the control group (from 61 ± 4 to 58 ± 12 mmol/mol; p = ns).

Table 3 Clinical and metabolic parameters throughout the study and changes (Δ) at the end of intervention in the study participants
Full size table

Both in the intervention and in the control groups, no change was recorded in the type and dosage of the drugs used for the control of diabetes and the other cardiovascular risk factors.

The data on dietary habits are available for 24 subjects of the control group and 59 subjects of the intervention group (Table 4). Both groups showed a significant reduction (200–300 kcal) in their daily energy intake, especially from fat. In particular, the intensive intervention group showed a greater reduction in saturated fat and a more marked increase in fibre intake than the control group (p < 0.009 and p < 0.0001, respectively); in contrast, the consumption of alcohol and mono-disaccharides decreased to a greater extent in the control than in the intervention group (p < 0.04 and p < 0.034, respectively).

Table 4 Eating habits throughout the study and changes (Δ) at the end of intervention in the study group
Full size table

Results at follow-up

At 9-month follow-up, all values except HDL cholesterol and arterial blood pressure returned to pre-intervention levels in control subjects (Table 3). Conversely, in the intervention group the reduction in body weight, waist circumference, glycated haemoglobin and blood pressure persisted over the follow-up (Table 3). During the follow-up, the energy and macronutrient intake of the control subjects did not substantially differ from the pre-intervention except for a persistent reduction in the intake of cholesterol and soluble carbohydrate (Table 4). In the intensive intervention group, the changes induced by the intervention persisted over the follow-up with regard to reduced energy, total and saturated fat, cholesterol intake and increased fibre consumption (Table 4).

During the follow-up, all participants were followed by their respective GP in a similar setting. Pharmacologic treatment for type 2 diabetes, hypertension and hypercholesterolaemia was maintained unchanged for the whole duration of the study, unless deemed necessary by the GP to avoid the risk of compromising patient’s health.

Discussion

On the whole, the composition of the diet followed by the study population at baseline deviates from the recommendations for a healthy diet, particularly with regard to the intake of total lipids, saturated fat and soluble carbohydrates, which were higher than the recommended levels, and the amount of fibre/1000 kcal which was below the recommended levels.

Short-term effects

The main finding of the study is the relevant and consistent change in metabolic parameters and cardiovascular risk factors obtained with the intensive dietary intervention. Indeed, a significant reduction in body weight, waist circumference and blood pressure also occurred in the control group, which demonstrates the efficacy of the health education programme. However, the extent of the clinical benefits was much greater in the intervention than in the control group, as evidenced by significant differences in body weight and waist circumference. In addition, the intervention group achieved a significant improvement in glucose control and lipid profile, which was not observed in the control group. These findings highlight the efficacy of the intensive dietary intervention in modifying the overall metabolic profile compared to the conventional treatment. Noteworthily, this intervention included an experiential approach that might have played a major role in changing the participants’ lifestyle in a sustained way, as demonstrated by the results of the follow-up.

From a nutritional point of view, an improvement in eating habits was observed after both conventional and intensive intervention. In the control group, the most interesting finding is related to the reduction in total energy intake, which is largely due to a lower fat consumption. In the intervention group, in addition to the reduction in energy and total fat intake, there was a decrease in saturated fatty acids and cholesterol. This finding suggests a lower consumption of animal products that are likely replaced by products of plant origin such as legumes, fruits and vegetables; this is confirmed by the increased fibre intake.

In short, in the control group there was an effort to modify eating habits, which had a little but significant effect on body weight but did not produce very relevant metabolic effects. In contrast, in the intervention group the combination of weight reduction and healthy dietary habits exerted a clear beneficial impact on the cardiometabolic risk profile.

By dividing the study population into subgroups on the basis of the prevailing disease (diabetes or hypertension), the observed differences between the two treatment groups were confirmed. As expected, the improvement in blood glucose was more marked in the subgroup of diabetic patients. Likewise, the decrease in blood pressure achieved in hypertensive patients was more pronounced than that in the overall population.

Follow-up data

In the intensive intervention group, an average 5-kg weight loss persisted over the follow-up confirming that the dietary intervention implemented in our study was more effective than the conventional one in achieving and maintaining weight loss along with healthy dietary habits with persistent improvement in cardiovascular risk factors.

The magnitude of the weight reduction documented in our study is comparable to that obtained in the Finnish Diabetes Prevention Study, where it was paralleled by a 58% reduction in the risk of diabetes in a population of subjects with IGT. On this basis, it is reasonable to assume that the intensive dietary intervention implemented in our study may have a clear clinical relevance [15].

The benefits achieved in the subgroups of both diabetic and hypertensive patients are particularly relevant since the reduction in glycated haemoglobin in diabetics and in blood pressure in hypertensive subjects is comparable to the effects obtained by pharmacologic treatments [16, 17] and has the potential to reduce the risk of cardiovascular events [17, 18].

With regard to the changes in eating habits, the intensive intervention group showed a significant decrease in energy intake as well as of total and saturated fatty acids and dietary cholesterol, as compared to baseline. These changes are known to exert beneficial effects on the atherosclerotic risk factor profile [19,20,21,22,23,24,25]. Noteworthily, these results were achieved by implementing a dietary pattern inspired to the traditional MD, which highlights the usefulness of this model as an educational tool, and easy to be proposed in view of its strong cultural and gastronomic roots.

Our study demonstrates that a lifestyle intervention, based not only on the transmission of theoretical concepts but also on participant’s experience, has long-lasting effects. In fact, the improvement in eating habits attained by the participants persisted for almost 1 year. It is worth considering that the benefits achieved were maintained over time in the absence of further interactions with the nutritionist in a setting where landmarks are, as for the general population, the GP and the diabetologist. Our study also supports the efficacy of cooking education programmes in improving people’s eating habits, thus confirming previous observations [26,27,28].

The major limitation of this study is the lack of randomization in assigning participants to the intensive or the conventional intervention group. However, random assignment was unfeasible because we could not apart from considering family- and work-related needs of the participants.

Conclusions

Based on the results obtained in this study, it seems appropriate to recommend the use of experiential methodologies to improve the lifestyle of population groups at increased metabolic and cardiovascular risk. In fact, a pivotal role in the achievement of our results is likely to be attributed to the methodology used. This was based not only on the traditional nutrition education but also on the consumption for a prolonged time of healthy meals prepared in compliance with the principles of the traditional MD, and on learning how to prepare meals according to the procedures typical of this nutritional model. By extending this methodology of intervention to school canteens and catering services at the worksites, it should be possible to broaden the number of potential beneficiaries of this type of preventive intervention. This approach could possibly improve population health and reduce the risk of major chronic non-communicable diseases that represent the major cause of death and disabilities in most countries.