Nutrition education in postmenopausal women: Changes in dietary and cardiovascular indices

Page 1

A

OpoMbCRiw(pCn©

K

1

G

f

0

Maturitas 55 (2006) 338–347

Nutrition education in postmenopausal women: Changesin dietary and cardiovascular indices

Yannis Manios ∗, George Moschonis, Evangelia Grammatikaki,Ioanna Katsaroli, Paraskevi Kanelou, Sofia Tanagra

Department of Nutrition and Dietetics, Harokopio University, 70, El.Venizelou Ave, 176 71 Kallithea, Athens, Greece

Received 7 January 2006; received in revised form 9 May 2006; accepted 16 May 2006

bstract

bjective: The aim of the current study was to examine whether a diet rich in dairy products followed by a nutrition educationrogram for the prevention of osteoporosis could have any adverse effect on certain cardiovascular disease (CVD) risk factorsver a 5-month intervention period.ethods: A total sample of 82 women (55–65 years old) was randomized to a dietary intervention group (IG: n = 42), attending

iweekly nutrition education program and provided with low-fat, fortified dairy products and to a control group (CG: n = 40).hanges in dietary, biochemical and clinical indices related to CVD were determined at the end of the 5-month intervention period.esults: The IG was found to have a higher decrease in the percentage of energy intake derived from total fat and a higher increase

n the intake of calcium, phosphorus, magnesium and potassium compared to the CG (p < 0.05). Furthermore, the IG subjectsere found to have a lower increase in BMI (0.7 ± 0.1 versus 1.4 ± 0.2 Kg/m2, p = 0.011) and systolic blood pressure (SBP)

2.5 ± 2.9 versus 7.8 ± 2.2 mmHg, p = 0.040) and a higher decrease in serum total cholesterol (−5.2 ± 3.3 versus 6.9 ± 5.1 mg/dl,

= 0.042) and LDL-cholesterol levels (−20.0 ± 2.6 versus −12.4 ± 4.2 mg/dl, p = 0.034) compared to the CG.onclusions: The findings of the current study indicate that a dietary intervention aiming to minimize the risk for osteoporosis didot have any adverse effects on CVD risk factors. On the contrary, it has induced favourable changes in BMI, serum lipids and SBP.

2006 Elsevier Ireland Ltd. All rights reserved.

lth Stud

eywords: Intervention; Diet; Blood pressure; Postmenopausal Hea

. Introduction

The seven-country study in 1960s indicated thatreece had the lowest mortality rates from cardio-

∗ Corresponding author. Tel.: +30 210 9549156;ax: +30 210 9549141.

E-mail address: [email protected] (Y. Manios).

vteauood

378-5122/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserdoi:10.1016/j.maturitas.2006.05.006

y; Serum lipids

ascular disease (CVD) compared to the other coun-ries participating in that first multi-centre nutritionalpidemiological study [1]. These low rates had beenttributed primarily to the dietary habits of the pop-

lation, which favoured a high fruit, vegetable andlive oil intake. However, before the end of the previ-us century Greece had one of the most rapidly risingeath rates due to CVD, which became the primary

ved.

Page 2

turitas

cbli[atoibmiitmsbddl

bpptshdGtodapcf

speioofsmtp

sp

2

2

taCubatKptwcasmTwoehcatoi

2

(s4nag

Y. Manios et al. / Ma

ause of morbidity and mortality [2,3]. This trend hadeen attributed to the adoption of a more “westernised”ifestyle combined with limited awareness on healthssues, poor dietary habits and the sedentary lifestyle3–6]. These lifestyle and dietary changes providedn explanation for the increased risk of CVD amonghe Greek adult population by the end of the previ-us century, since a large proportion of the energyntake derived from fruits and vegetables in 1960s haseen replaced by other products, such as dairy andeat [5,7]. Among these foods, those widely consumed

n Greece are quite energy dense and provide a highntake of saturated fat and sodium, which explains whyhey are considered as risk factors for the develop-ent of obesity, hypercholesterolemia and hyperten-

ion. Baring all these in mind an opposite trend haseen observed among health professionals over the lastecade encouraging the general public in Greece toecrease intake of meat and dairy products in order toower the risk for CVD.

However, low consumption of dairy products haseen also associated with an increased risk for osteo-orosis [8]. Since life expectancy is increasing therevalence of osteoporosis is rising accordingly, par-icularly among postmenopausal women, reflecting aignificant burden of morbidity and mortality, whichas become a major public health problem in nowa-ays [9]. Recent epidemiological studies conducted inreece over the last few years have recorded a different

rend among postmenopausal women regarding intakef dairy products. Compared to what would be expecteduring the previous decade, intake of dairy productsmong this age group has decreased while an increasedrevalence of osteopenia and osteoporosis [10] andonsequently an increased risk for osteoporotic boneracture has been recorded [11].

These contradicting findings regarding dairy con-umption and the increased risk for CVD and osteo-orosis [12] highlight the confusion among the gen-ral public as well as health professionals regard-ng recommended intakes of dairy products. The aimf the current study was to examine the changesn dietary indices but also on certain CVD riskactors, namely BMI, serum lipids and blood pres-

ure on a postmenopausal population following a 5-onth nutrition education program for the preven-

ion of osteoporosis. In addition to the interventionrogram, subjects were provided and advised to con-

Tm

I

55 (2006) 338–347 339

ume daily three portions of fortified, low fat dairyroducts.

. Methods

.1. Sampling

The recruitment of the study participants followedwo screenings, conducted in the period between Julynd November 2004, after the approval of the Ethicalommittee of Harokopio University of Athens. Vol-nteers’ recruitment was conducted via informationalrochures and posters distributed in public buildingsnd community centers in three municipalities withinhe wider district of Athens, namely Nea Smyrni,allithea and Neo Iraklio. Through this process 307ostmenopausal women registered voluntarily to par-icipate in the study. Following this first call all subjectsere asked to go through a screening including medi-

al history, demographic data, dietary, physical activitynd smoking habits, as well as bone mineral den-ity assessment by calcaneal quantitative ultrasoundeasurement and dual-energy X-ray absorptiometry.his screening finally yielded 82 eligible women, whoere not diagnosed having osteoporosis or any kindf bone and cardiovascular degenerative chronic dis-ase (i.e. diabetes, heart disease, cancer, hyper- andypothyroidism etc.), were not routinely taking medi-ations and supplements, were non-smokers and weret menopause for more than 3 years. Prior to their entryo the study the subjects signed a detailed consent formf participation and proceeded to the next phase (thentervention phase) of the study.

.2. Study groups and intervention scheme

These 82 self-dependent postmenopausal womenage 55–65 years old) were randomly assigned into twotudy groups. The intervention group (IG) comprised of2 women who were invited to participate in biweeklyutrition education sessions of approximately 1 h eachnd the remaining 40 women comprised the controlroup (CG) to whom no intervention was delivered.

he intervention period had a total duration of about 5onths (November 2004–March 2005).The aim of the biweekly sessions provided for the

G was to increase awareness of the subjects on health

Page 3

3 turitas

ivtsefordtiirdibilcHjtsec7

aodav

2

tatfa

2

idi

vtdfsrhfduDafaG

uw(sc(

A

SimitrSd

2

sups

40 Y. Manios et al. / Ma

ssues, primarily related to osteoporosis, but also moti-ate them to change certain lifestyle and dietary pat-erns in order to improve their health status. The firstessions, with the use of power point presentations andasy to understand terminology and slides, primarilyocused in educating the subjects on the pathophysi-logy of the disease as well as the risk factors (healthelated behaviours) related to the development of theisease. Gradually the sessions became more interac-ive and emphasis was placed in guiding and assist-ng the subjects in changing their dietary habits andncreasing their weekly physical activity. Particularlyegarding dietary habits more emphasis was given inecreasing the fat intake of the subjects and increas-ng the intake of micronutrients primarily related toone metabolism such as calcium and vitamin D. Toncrease compliance, IG subjects were supplied withow fat dairy products (milk and yogurt) fortified withalcium and vitamin D (provided by Friesland Foodsellas). In order to avoid excess caloric intake IG sub-

ects were advised to substitute other dairy products inheir diet with those provided, while the new dietarycheme they were advised to follow was low in fat butfficient in providing the recommended daily intake ofalcium and vitamin D (about 1200 mg of Calcium and.5 mg of vitamin D per day).

The CG subjects, after completion of both baselinend follow-up examinations, were given the outcomef their medical screening presented below including aiagnosis of the medical staff of the research team. Nodditional health and nutrition intervention was pro-ided for the CG subjects.

.3. Assessment of the effects of the intervention

In order to assess the effects of the intervention cer-ain behavioural and clinical indices were obtained botht baseline and follow-up. The baseline examinationook place during October–November 2004, while theollow-up examination took place in March 2005 andfter almost 5 months of intervention.

.4. Dietary indices

The 24 h recall technique was used to collect dietaryntake information for a total of 3 days, 2 week-ays and 1 weekend day, most preferably Sunday. Allnterviewers were rigorously trained to minimize inter-

smB(

55 (2006) 338–347

iewer’s effect. Respondents were asked to recall theype and amount of any food and beverage consumeduring the previous day in a chronological order, i.e.rom the time they woke up in the morning to theame time the following day. To improve the accu-acy of the reported food quantities, standard house-old measures (cups, tablespoons, etc.) and pictureood models (Dairy Food Council, USA) were useduring interviews. Food intake data were analyzedsing the Nutritionist V diet analysis software (Firstatabank, San Bruno, CA), which was extensively

mended to include food composition tables [13,14]or traditional Greek foods and recipes and chemicallynalyzed commercial food items widely consumed inreece.The distribution of usual intakes was estimated by

sing the National Research Council method [15,16],hich attempts to remove the effects of day-to-day

within-subject) and the subject-by-subject (betweenubjects’) variability in dietary intakes. More specifi-ally, the equation used for the calculation of adjustedusual) dietary intake was the following:

djusted Intake = [(subject’s mean − group mean)

×S.D.between-person

S.D.observed] + group mean

ubject’s mean refers to each subject’s mean dietaryntake as obtained by the 3-day dietary recalls. Group

ean refers to the studied population’s mean dietaryntake. S.D.between-person refers to the square root ofhe variance of the distribution of usual intakes, cor-ected for the effects of the within subject variability..D.observed refers to the square root of variance of theistribution of the reported intakes.

.5. Anthropometrical indices

Anthropometry was carried out during the initialcreening, as well as during the baseline and the follow-p re-examinations. In all the aforementioned timeoints body weight and standing height were mea-ured in light clothing and with no shoes using a digital

cale (Seca) with an accuracy of ±100 g and a com-ercial stadiometer to the nearest 0.5 cm, respectively.ody mass index (BMI) was calculated as weight

kg) divided by height squared (m2). The rest of the

Page 4

turitas

afasloetbtt[

2

fiPmvawstUt

cdat(cerp(dm

2

bmta

frtae

2

sptsbngoyttiStc

3

t5fnfT(ftptfftl

Y. Manios et al. / Ma

nthropometrical indices were assessed at baseline andollow-up examinations following the same proceduresnd using the same standardized equipments. Morepecifically, waist circumference was measured at theevel of umbilicus and hip circumference at the levelf greater trochanters and pubic symphisis to the near-st 0.1 cm. Left arm circumference was measured athe mid-upper-arm point, which is half the distanceetween the acromion and the olecranon. Finally leftriceps, biceps, subscapular and suprailiac skinfoldhickness was measured with a Lange skinfold caliper17].

.6. Biochemical indices

Early-morning venous blood samples were obtainedrom each subject for biochemical and haematolog-cal screening tests, following a 12 h overnight fast.rofessional staff performed venipuncture to obtain aaximum of 25 ml blood. The blood was collected in

acutainers with no added anticoagulant and was keptt room temperature for approximately 2 h, where itas allowed to clot as this was designated for serum

eparation. Biochemical analyses were conducted athe Nutrition and Metabolism Laboratory of Harokopioniversity, following centrifugation for serum separa-

ion at 3000 rpm for 15 min.Total cholesterol (TC), high-density lipoprotein

holesterol (HDL-C), and triglycerides (TG) wereetermined in duplicate using commercially avail-ble enzymatic colorimetric assays (Roche Diagnos-ics GmbH, Mannheim) on an automated analyzerRoche/Hitachi Modular). Low-density lipoproteinholesterol (LDL-C) was calculated by the Friedewaldquation [18]. The LDL-C/HDL-C and the TC/HDL-Catios were also estimated. Serum calcium and serumhosphorus were determined by a colorimetric assayRoche Diagnostics GmbH, Mannheim) with endpointetermination and sample blank [19,20] on an auto-ated analyzer (Roche/Hitachi Modular).

.7. Blood pressure measurements

Blood pressure (BP) measurements were performed

y a trained physician using a standard mercury sphyg-omanometer. BP was measured on the right arm, with

he subject seated and quiet after a 5 min rest, the fore-rm at heart level, and the stethoscope in the antecubital

Tdtg

55 (2006) 338–347 341

ossa. Systolic (SBP) and diastolic BP (DBP) wereecorded as the first and fifth Korotkoff phases, respec-ively. If the pressure readings differed by 10 mmHg,

third reading was taken. The mean reading at eachxamination was used in data analysis.

.8. Statistical analysis

Continuous variables were reported as mean ±tandard error (S.E.) and categorical variables as pro-ortions (%). The unpaired Student’s t-test was usedo evaluate any possible differences on baseline mea-urements between those women with participation inoth baseline and follow-up examinations and thoseot re-examined at follow-up. Differences betweenroups regarding proportions were evaluated using χ2

r Fisher’s exact test. General linear mixed model anal-sis was used to evaluate the differences between thewo groups regarding the changes observed for cer-ain anthropometrical, biochemical and blood pressurendices, after controlling for potential confounders.PSS 11.0 (SPSS Inc. Texas, USA) software was used

o conduct all statistical analyses. The level of statisti-al significance was set at p ≤ 0.05.

. Results

Seven of the 82 women initially assigned to par-icipate in the study could not be re-examined at the-month follow-up providing a total of 75 women withull data from both examinations. Consequently theumber of subjects in each group with full baseline andollow-up data was 39 for the IG and 36 for the CG.he mean age of these women was 60.0 ± 4.8 years

age range 55–65 years). No statistically significant dif-erences were observed regarding the mean values ofhe baseline anthropometrical, biochemical and bloodressure data investigated in the current study betweenhose women with participation in both baseline andollow-up examinations and those not re-examined atollow-up (data not shown). Baseline characteristics ofhe 75 study participants with full data in both base-ine and follow-up examinations are summarized in

able 1. No significant differences were observed inemographic and anthropometrical indices between thewo groups examined, thus indicating a relative homo-eneity.

Page 5

342 Y. Manios et al. / Maturitas 55 (2006) 338–347

Table 1Comparison of baseline demographic and anthropometrical characteristics between intervention (n = 39) and control (n = 36) group

Intervention group, mean ± S.E. Control group, mean ± S.E. p-Value†

Age (years) 60.5 ± 0.71 61.4 ± 0.85 0.074Years since menopause 9.2 ± 0.9 10.4 ± 1.1 0.146Weight (Kg) 71.2 ± 1.5 74.9 ± 2.0 0.144Height (cm) 158.9 ± 1.1 156.9 ± 1.0 0.159

Educational level % %

Up to 9 years 53.8 62.2 0.4239–12 years 28.2 23.9More than 12 years 17.9 13.9Marital status

Married 64.1 69.4 0.807Single/divorced/widows 35.9 30.6

s evaluav

cchdHofsjtta0Rhv(mp2wtg

Bh0bfs

os(fcfof2to

4

siiwolsassd

† The level of statistical significance for continuous variables waariables with the use of χ2 and Fisher’s exact test.

According to the data presented in Table 2 thehanges observed for energy intake and for the per-ent of energy derived from the intake of carbo-ydrates, saturated and monounsaturated fatty acids,id not differ significantly between the two groups.owever the increase with respect to the percentf energy derived from protein intake was higheror the IG compared to the CG (4.5 ± 0.6 ver-us 1.0 ± 0.8%, p = 0.001). Furthermore the IG sub-ects were found to have a higher decrease inhe percent of their energy intake derived fromotal fat (−2.1 ± 0.9 versus −0.2 ± 0.8%, p = 0.048)nd polyunsaturated fatty acids (−0.9 ± 0.4 versus.2 ± 0.4%, p = 0.046) compared to the CG ones.egarding micronutrients the IG was found to exhibit aigher increase in the intake of calcium (566.8 ± 94.5ersus 20.1 ± 62.5 mg/day, p < 0.001), phosphorus406.8 ± 87.0 versus 60.7 ± 68.1 mg/day, p = 0.003),agnesium (100.1 ± 14.2 versus 32.1 ± 10.0 mg/day,= 0.001) and potassium (385.3 ± 112.7 versus7.6 ± 107.0 mg/day, p = 0.027) compared to the CG,hereas the change in sodium intake was not found

o differentiate significantly between the two studyroups.

As Table 3 illustrates the increase observed forMI during the intervention period was significantlyigher for the CG compared to the IG (1.4 ± 0.2 versus

.7 ± 0.1 Kg/m2, p = 0.011), while the decrease overaseline for sum of skinfolds was found to be higheror the IG, compared to the CG (−16.4 ± 3.5 ver-us −5.7 ± 4.1 mm, p = 0.036). Similarly, the decrease

ab

t

ted with the use of unpaired Student’s t-test, while for categorical

bserved for the IG in serum TC (−5.2 ± 3.3 ver-us 6.9 ± 5.1 mg/dl, p = 0.042) and LDL-C levels−20.0 ± 2.6 versus −12.4 ± 4.2 mg/dl, p = 0.034) wasound to be significantly higher than the respectivehanges in these two biochemical indices observedor the CG. Regarding blood pressure the increasebserved for SBP was found to be significantly higheror the CG, compared to the IG (7.8 ± 2.2 versus.5 ± 2.9 mmHg, p = 0.040). There were no other sta-istically significant differences, with respect to the restf the indices examined.

. Discussion

The findings of the current study have indicatedome significant favourable changes in micronutrients’ntake for the IG subjects (Table 2). More specif-cally calcium, phosphorus and magnesium intakesere favourably increased for the IG, reaching the rec-mmended or close to the recommended daily intakeevels [16]. Still that was not observed for potassium,ince its daily intake was below recommendations,lthough its increase from baseline to follow-up wasignificantly higher for the IG than the CG. Similarlyodium intake was below the recommended level, butue to poor reported data on added salt during cooking

nd food preparation, this particular finding needs toe treated with caution.

The changes observed in the intake of the aforemen-ioned micronutrients should primarily be attributed to

Page 6

Y. Manios et al. / Maturitas 55 (2006) 338–347 343

Table 2Comparisons of changes in daily energy, macro- and micro-nutrient intakes between intervention (n = 39) and control (n = 36) group followingthe 5-month intervention period

Baseline (November2004) (mean ± S.E.)

Follow-up (April2005) (mean ± S.E.)

5-month change(mean ± S.E.)

p-Value*

Energy and macronutrients intakeEnergy (Kcal/day)

Intervention group 1622.5 ± 52.6 1596.6 ± 58.9 −26.0 ± 65.1 0.427Control group 1708.2 ± 54.6 1623.7 ± 61.2 −84.6 ± 67.6

Protein (% of energy intake)Intervention group 15.8 ± 0.4 20.3 ± 0.5 4.5 ± 0.6 0.001Control group 15.4 ± 0.5 16.4 ± 0.7 1.0 ± 0.8

Carbohydrate (% of energy intake)Intervention group 43.9 ± 1.4 43.3 ± 0.6 -0.6 ± 1.3 0.116Control group 43.6 ± 1.2 45.7 ± 1.3 2.1 ± 1.5

Total fat (% of energy intake)Intervention group 38.7 ± 1.0 36.6 ± 0.8 −2.1 ± 0.9 0.048Control group 38.9 ± 1.0 38.7 ± 0.9 −0.2 ± 0.8

Saturated fatty acids (% of energy intake)Intervention group 11.2 ± 0.3 11.8 ± 0.2 0.6 ± 0.2 0.119Control group 11.4 ± 0.4 11.3 ± 0.4 −0.1 ± 0.5

Monounsaturated fatty acids (% of energy intake)Intervention group 20.5 ± 0.6 18.7 ± 0.6 −1.8 ± 0.7 0.126Control group 20.4 ± 0.7 20.1 ± 0.5 −0.3 ± 0.7

Polyunsaturated fatty acids (% of energy intake)Intervention group 7.0 ± 0.4 6.1 ± 0.2 −0.9 ± 0.4 0.046Control group 7.1 ± 0.4 7.3 ± 0.3 0.2 ± 0.4

Micronutrients intakeSodium (mg/day)

Intervention group 2076.2 ± 147.3 1874.2 ± 139.4 −202.0 ± 146.7 0.924Control group 2033.0 ± 150.7 1849.1 ± 147.2 −183.8 ± 165.9

Potassium (mg/day)Intervention group 2054.1 ± 79.8 2439.4 ± 126.9 385.3 ± 112.7 0.027Control group 2047.4 ± 86.8 2075.0 ± 98.1 27.6 ± 107.0

Calcium (mg/day)Intervention group 681.6 ± 33.6 1248.5 ± 100.2 566.8 ± 94.5 <0.001Control group 709.7 ± 40.0 729.8 ± 61.5 20.1 ± 62.5

Phosphorus intake (mg/day)Intervention group 974.8 ± 37.8 1381.6 ± 94.1 406.8 ± 87.0 0.003Control group 942.7 ± 47.8 1003.4 ± 62.6 60.7 ± 68.1

Magnesium (mg/day)Intervention group 192.1 ± 5.4 292.2 ± 15.1 100.1 ± 14.2 0.001

226.

evaluatf

tItirn(d

afot

Control group 194.1 ± 5.8

* p-Values regarding the differences between the two groups wereor changes in BMI.

he increased consumption of dairy products by theG. This also explains the increased intake of pro-ein observed for the IG, since no increase in meatntake or other foods with high protein content were

eported. However, the increase in protein intake wasot followed by an increase in the mean energy intakeTable 2). This caloric stability observed for the IGuring the intervention period seems to be mainly

fucr

2 ± 9.4 32.1 ± 10.0

ed by using linear mixed models analysis. Adjustments were made

ttributed to the decrease in energy derived from totalat, primarily due to a restriction in the consumptionf margarine, butter and desserts with high fat con-ent. Still a significant proportion of this decline in

at intake was also due to the reduction of olive oilsed in salads and general food preparation. Olive oilonstitutes a large proportion of fat intake in Mediter-anean populations and although its fatty acid content

Page 7

344 Y. Manios et al. / Maturitas 55 (2006) 338–347

Table 3Comparisons of changes in certain CVD risk factors between intervention (n = 39) and control (n = 36) group following the 5-month interventionperiod

Baseline (November2004) (mean ± S.E.)

Follow-up (April2005) (mean ± S.E.)

5-month change(mean ± S.E.)

p-Value*

Anthropometrical indicesBody mass index (Kg/m2)

Intervention group 28.3 ± 0.6 28.9 ± 0.7 0.7 ± 0.1 0.011Control group 29.8 ± 0.9 31.2 ± 0.9 1.4 ± 0.2

Sum of skinfolds (mm)Intervention group 134.2 ± 5.1 117.8 ± 4.2 −16.4 ± 3.5 0.036Control group 141.3 ± 5.7 135.6 ± 5.5 −5.7 ± 4.1

Waist to hip ratioIntervention group 0.85 ± 0.01 0.80 ± 0.01 −0.05 ± 0.01 0.781Control group 0.87 ± 0.01 0.82 ± 0.01 −0.05 ± 0.01

Biochemical indicesTotal cholesterol (mg/dl)

Intervention group 226.8 ± 5.5 221.6 ± 5.5 −5.2 ± 3.3 0.042Control group 215.1 ± 4.6 222.0 ± 5.4 6.9 ± 5.1

HDL-cholesterol (mg/ml)Intervention group 62.2 ± 1.8 60.9 ± 1.9 −1.4 ± 0.8 0.563Control group 60.9 ± 2.1 60.8 ± 1.9 −0.1 ± 1.0

LDL-cholesterol (mg/ml)Intervention group 141.7 ± 3.7 121.6 ± 4.0 −20.0 ± 2.6 0.034Control group 133.1 ± 3.5 120.7 ± 4.3 −12.4 ± 4.2

Total cholesterol: HDL-cholesterolIntervention group 3.76 ± 0.14 3.80 ± 0.18 0.04 ± 0.09 0.690Control group 3.65 ± 0.13 3.75 ± 0.13 0.10 ± 0.11

LDL-cholesterol: HDL-cholesterolIntervention group 2.37 ± 0.11 2.09 ± 0.13 −0.28 ± 0.07 0.775Control group 2.26 ± 0.11 2.04 ± 0.11 −0.22 ± 0.11

Triglycerides (mg/ml)Intervention group 110.6 ± 7.5 119.4 ± 8.6 8.7 ± 5.2 0.177Control group 107.5 ± 8.5 124.0 ± 8.7 16.5 ± 5.9

Blood pressure indicesSystolic blood pressure (mmHg)

Intervention group 131.0 ± 2.7 133.5 ± 3.1 2.5 ± 2.9 0.040Control group 136.4 ± 3.4 144.3 ± 4.1 7.8 ± 2.2

Diastolic blood pressure (mmHg)Intervention group 81.5 ± 1.6 83.7 ± 1.3 2.2 ± 1.6 0.310Control group 81.7 ± 1.7 83.5 ± 1.5 1.8 ± 1.6

evaluatf

iihas

p

* p-Values regarding the differences between the two groups wereor age and for changes in BMI.

s much more favourable compared to other fats, still

ts excessive consumption partially contributes to theigh prevalence of obesity reported in Greece [7,21]nd other Mediterranean countries [22]. For this rea-on the subjects in the IG were advised to measure the

ttti

ed by using linear mixed models analysis. Adjustments were made

ortions of the olive oil used in food preparation and

o be aware of its high energy content. In consistencyo the current findings, several other dietary interven-ion trials have reported respective favourable changesn fat intake among postmenopausal women [23–25],

Page 8

turitas

tce

Bsfwlplw(iiIeiaciconbmbfw

aiLmsmwfatlifCtlrf

eeresdtscbtfIicemtbwtctsbrdetisaats

dpoipom

Y. Manios et al. / Ma

hus highlighting the effectiveness of nutrition edu-ation in producing and maintaining health-orientedating behaviours and practices.

Regarding anthropometrical indices the increase inMI observed for both study groups could possibly

tem from the seasonal variation in weight reportedor the overall population and particularly for women,ho seem to gain weight during winter months and

ose it during the summer period [26,27]. Anotherossible explanation could be the considerably lowevels of physical activity recorded for all subjects,hich did not change during the intervention period

data not presented). However according to the find-ngs of the current study a significantly lower increasen BMI was found for the IG, compared to the CG.n turn this observation should be attributed to theffectiveness of the intervention program in motivat-ng women and in improving their self-efficacy indopting and maintaining a more balanced diet. Thisould also provide some support for the greater declinen the IG subjects’ subcutaneous fat mass as indi-ated by the sum of skinfolds. Still the inconsistencybserved for the changes obtained for skinfold thick-ess and BMI over the intervention period could alsoe attributed to a lower reliability of the skinfolds’easurement procedure. Analogous findings have

een reported by the “Women’s Health Trial” studyor a respective population group of postmenopausalomen [28].The aforementioned changes observed in dietary

nd anthropometrical indices could probably havenduced the favourable changes observed in TC andDL-C levels for the IG. These findings are in agree-ent with other recent studies, which have reported

imilar changes in serum lipid levels after the imple-entation of nutritional intervention trials amongomen [24,29,30]. The significant changes observed

or TC and LDL-C during follow-up examination couldlso be attributed to the unfavourably high levels ofhese indices obtained for both groups at baseline, thuseaving space for significant improvements during thentervention. On the other hand, the changes observedor the rest of the serum lipids examined (i.e. HDL-

and TG) were not found to differentiate between the

wo study groups, probably due to the fact that baselineevels of these indices were already within the normalange for both groups and therefore there was no roomor any further improvement.

easw

55 (2006) 338–347 345

Similarly to BMI, increases in SBP and DBP lev-ls were observed for both IG and CG. Most researchmphasizes on the fact that blood pressure is positivelyelated to obesity as well as to abnormally high lev-ls of certain biochemical indices such as TC, TG anderum insulin [31–33], which are further affected byietary habits and other behavioural factors. However,he increase in SBP observed in the current study wasignificantly lower for the IG. The difference in SBPhanges between the two groups could be attributedoth to the significantly higher increase of BMI inhe CG compared to the IG as well as to the moreavourable changes in micronutrients’ intake in theG compared to the CG. More specifically, the higherncrease in the intake of potassium, magnesium andalcium by the IG is in line with the available lit-rature, thus highlighting the protective role of theseicronutrients against abnormal blood pressure eleva-

ion [34,35]. In general dietary modifications shoulde strongly encouraged in postmenopausal womenho have blood pressure levels within the prehyper-

ensive and hypertensive range. This modificationsould involve portion-size control, decreased consump-ion of sugar-containing beverages and energy-densenacks, increased consumption of fresh fruits, vegeta-les and low-fat dairy products and introduction ofegular meals, including a healthy breakfast, in theiraily dietary scheme [36]. All these dietary goals werencouraged through the implemented nutrition educa-ion program and as revealed by the observed dietaryntake changes, most were actually met by the IGubjects. Still, sodium intake changes could not beccurately recorded due to the poor reported data ondded salt during cooking and food preparation andhat could be considered a limitation of the currenttudy.

The findings of the current study indicate that aietary scheme aiming to minimize the risk for osteo-orosis does not counteract CVD prevention. On thepposite, supporting postmenopausal women in adopt-ng a balanced diet comprising low-fat dairy products,lenty of fruits and vegetables and a prudent use oflive oil seems to be a promising approach in mini-izing the risk of both chronic diseases. Although the

ffectiveness of any intervention program can only bessessed in the long-term, the findings of the currenttudy provide some directions for health professionalsorking in the field of chronic disease prevention.

Page 9

tas

Unbalanced diet to lower serum cholesterol level is a risk factorfor postmenopausal osteoporosis and distal forearm fracture.Osteoporos Int 2001;12:296–301.

[13] University of Crete: Food Composition Tables. nutri-tion.med.uoc.gr/GreekTables, 1991.

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

55 (2006) 338–347

14] Trichopoulou A. Composition tables of foods and Greek dishes.Athens, Greece: School of Medicine: Department of Hygieneand Epidemiology; 2004.

15] National Research Council. Nutrient adequacy: assessmentusing food consumption surveys. Report of the food and nutri-tion board. Washington, DC: National Academy of Sciences;1986.

16] Institute of Medicine Food and Nutrition Board. Dietary refer-ence intakes: applications in dietary assessment. Washington,DC: National Academy Press; 2001.

17] Lohman T, Roche A, Mastorell R. Anthropometric standardiza-tion reference manual. Champaign, IL: Human Kinetics; 1991.

18] Friedewald WT, Levy RI, Fredrickson DS. Estimation of theconcentration of low-density lipoprotein cholesterol in plasma,without use of the preparative ultracentrifuge. Clin Chem1972;18:499–502.

19] Gindler EM, King JD. Rapid colorimetric determination of cal-cium in biologic fluids with methylthymol blue. Am J ClinPathol 1972;58:376–82.

20] Henry R. Clinical chemistry: principles and technics. 2nd ed.New York, NY: Harper & Row; 1974.

21] Manios Y, Panagiotakos DB, Pitsavos C, Polychronopoulos E,Stefanadis C. Implication of socio-economic status on the preva-lence of overweight and obesity in Greek adults: the ATTICAstudy. Health Policy 2005;74:224–32.

22] Popkin BM, Gordon-Larsen P. The nutrition transition: world-wide obesity dynamics and their determinants. Int J Obes RelatMetab Disord 2004;28(Suppl 3):S2–9.

23] Bowen D, Ehret C, Pedersen M, et al. Results of an adjunctdietary intervention program in the Women’s Health Initiative.J Am Diet Assoc 2002;102:1631–7.

24] Goulet J, Lamarche B, Nadeau G, Lemieux S. Effect of a nutri-tional intervention promoting the Mediterranean food pattern onplasma lipids, lipoproteins and body weight in healthy French-Canadian women. Atherosclerosis 2003;170:115–24.

25] Bhargava A, Hays J. Behavioral variables and education arepredictors of dietary change in the Women’s Health Trial: feasi-bility study in minority populations. Prev Med 2004;38:442–51.

26] Bellisle F, Dalix A, de Castro JM. Eating patterns in Frenchsubjects studied by the “weekly food diary” method. Appetite1999;32:46–52.

27] Sasaki T, Sakamoto K, Akaho R, Nakajima T, TakahashiK. Familial transmission of seasonal changes in sleep andeating function in the general population. Psychiatry Res1998;81:211–7.

28] Bhargava A, Guthrie JF. Unhealthy eating habits, physical exer-cise and macronutrient intakes are predictors of anthropometricindicators in the Women’s Health Trial: feasibility study inminority populations. Br J Nutr 2002;88:719–28.

29] Rock CL, Flatt SW, Thomson CA, et al. Plasma triacylglyc-erol and HDL cholesterol concentrations confirm self-reportedchanges in carbohydrate and fat intakes in women in a diet

346 Y. Manios et al. / Maturi

Acknowledgement

The study was supported by a research grant fromFriesland Foods Hellas. The authors would like to thankMaria Skotsimara, Anna Nikolaidi, Kostantinos Kout-sikas, dietitians and Katerina Tsikardoni, physical edu-cation instructor, for their valuable contribution in datacollection and processing.

References

[1] Kromhout D, Keys A, Aravanis C, et al. Food consumptionpatterns in the 1960s in seven countries. Am J Clin Nutr1989;49:889–94.

[2] Pitsavos C, Panagiotakos DB, Chrysohoou C, Stefanadis C. Epi-demiology of cardiovascular risk factors in Greece: aims, designand baseline characteristics of the ATTICA study. BMC PublicHealth 2003;3:32.

[3] Kafatos A, Papoutsakis G. Mortality rates in Greece and theirrelationship to the Mediterranean diet and ti health and nutritioneducation. Iatriki 1998;73:287–301.

[4] Kafatos A, Manios Y, Markatji I, Giachetti I, Vaz de AlmeidaMD, Engstrom LM. Regional, demographic and national influ-ences on attitudes and beliefs with regard to physical activity,body weight and health in a nationally representative sam-ple in the European Union. Public Health Nutr 1999;2:87–95.

[5] Kafatos A, Mamalakis G. Changing patterns of fat intake inCrete. Eur J Clin Nutr 1993;47(Suppl 1):S21–4.

[6] Voukiklaris GE, Kafatos A, Dontas AS. Changing prevalence ofcoronary heart disease risk factors and cardiovascular diseasesin men of a rural area of Crete from 1960 to 1991. Angiology1996;47:43–9.

[7] Ferro-Luzzi A, James WP, Kafatos A. The high-fat Greek diet:a recipe for all? Eur J Clin Nutr 2002;56:796–809.

[8] Heaney RP. Calcium, dairy products and osteoporosis. J AmColl Nutr 2000;19:83S–99S.

[9] Hodgson SF, Watts NB, Bilezikian JP, et al. American Associa-tion of Clinical Endocrinologists medical guidelines for clinicalpractice for the prevention and treatment of postmenopausalosteoporosis: 2001 edition, with selected updates for 2003.Endocr Pract 2003;9:544–64.

[10] Bletsa M, Babaroutsi E, Sidossis S, Manios Y, Sidossis LS.Prevalence of osteopenia and osteoporosis in Greece. In: 2ndCongress of the European Chapter of the American College ofNutrition. Athens; November 15–16, 2002.

[11] Paspati I, Galanos A, Lyritis GP. Hip fracture epidemiology inGreece during 1977–1992. Calcif Tissue Int 1998;62:542–7.

[12] Varenna M, Binelli L, Zucchi F, Ghiringhelli D, Sinigaglia L.

intervention trial. J Nutr 2004;134:342–7.

30] Stoddard AM, Palombo R, Troped PJ, Sorensen G,Will JC. Cardiovascular disease risk reduction: the Mas-sachusetts WISEWOMAN project. J Womens Health (Larchmt)2004;13:539–46.

Page 10

turitas

[

[

[

[

[

Y. Manios et al. / Ma

31] Burke V, Beilin LJ, Dunbar D. Tracking of blood pres-sure in Australian children. J Hypertens 2001;19:1185–92.

32] Raitakari OT, Porkka KV, Rasanen L, Viikari JS. Relations oflife-style with lipids, blood pressure and insulin in adolescents

and young adults. The cardiovascular risk in Young Finns Study.Atherosclerosis 1994;111:237–46.

33] Yong LC, Kuller LH. Tracking of blood pressure from adoles-cence to middle age: the Dormont High School Study. Prev Med1994;23:418–26.

[

55 (2006) 338–347 347

34] Lancaster KJ, Smiciklas-Wright H, Weitzel LB, Mitchell DC,Friedmann JM, Jensen GL. Hypertension-related dietary pat-terns of rural older adults. Prev Med 2004;38:812–8.

35] Zhou BF, Stamler J, Dennis B, et al. Nutrient intakes of middle-aged men and women in China, Japan, United Kingdom, and

United States in the late 1990s: the INTERMAP study. J HumHypertens 2003;17:623–30.

36] Nowson CA, Worsley A, Margerison C, et al. Blood pressureresponse to dietary modifications in free-living individuals. JNutr 2004;134:2322–9.