Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Does the Mediterranean diet predict longevity in the elderly? A Swedish perspective

Does the Mediterranean diet predict longevity in the elderly? A Swedish perspective AGE (2011) 33:439–450 DOI 10.1007/s11357-010-9193-1 Does the Mediterranean diet predict longevity in the elderly? A Swedish perspective Gianluca Tognon & Elisabet Rothenberg & Gabriele Eiben & Valter Sundh & Anna Winkvist & Lauren Lissner Received: 14 July 2010 /Accepted: 28 October 2010 /Published online: 26 November 2010 The Author(s) 2010. This article is published with open access at Springerlink.com Abstract Dietary pattern analysis represents a useful censoring at fixed follow-up time, adjusting for improvement in the investigation of diet and health activities of daily living and main cardiovascular risk relationships. Particularly, the Mediterranean diet factors including weight/waist circumference changes pattern has been associated with reduced mortality at follow up. In conclusion, we can reasonably state risk in several studies involving both younger and that a higher adherence to a Mediterranean diet pattern, elderly population groups. In this research, relation- especially by consuming wholegrain cereals, foods ships between dietary macronutrient composition, as rich in polyunsaturated fatty acids, and a limited well as the Mediterranean diet, and total mortality amount of alcohol, predicts increased longevity in the were assessed in 1,037 seventy-year-old subjects (540 elderly. females) information. Diet macronutrient composition . . . was not associated with mortality, while a refined Keywords Elderly Mediterranean diet Diet quality . . version of the modified Mediterranean diet index Macronutrient intake Diet adherence Mortality showed a significant inverse association (HR=0.93, 95% CI: 0.89; 0.98). As expected, inactive subjects, smokers and those with a higher waist circumference Introduction had a higher mortality, while a reduced risk charac- terized married and more educated people. Sensitivity Epidemiologic studies indicate that quality of diet analyses (which confirmed our results) consisted of: together with other lifestyle factors have considerable exclusion of one food group at a time in the influence on health status as well as mortality risk Mediterranean diet index, exclusion of early deaths, (Huijbregts et al. 1997; Seymour et al. 2003; Haveman-Nies et al. 2003). Although many studies have examined effects of single nutrients, foods, or food : : : G. Tognon (*) G. Eiben V. Sundh L. Lissner groups on health status, it is becoming increasingly Public Health Epidemiology Unit, evident that the assessment of dietary patterns could be a Department of Public Health and Community Medicine, practical alternative in the study of diet and health University of Gothenburg, Sahlgrenska Academy, associations (Mai et al. 2005). Among other a priori Box 454, 405 30 Göteborg, Sweden e-mail: [email protected] patterns (such as the Healthy Eating Index or the Recommended Food Score), the Mediterranean diet E. Rothenberg A. Winkvist has already been shown to possess many beneficial Department of Clinical Nutrition, effects, as first became evident years ago in the Seven- University of Gothenburg, Sahlgrenska Academy, Box 459, 405 30 Göteborg, Sweden Country Study (Keys 1980). This diet is characterized 440 AGE (2011) 33:439–450 by a high intake of vegetables, legumes, fruits, nuts while the 70-year-old women born in those years and seeds, cereals (that in the past were largely were identified on the basis of previous inclusion in unrefined), and a high intake of olive oil combined the Prospective Population Study of Women in with a low intake of saturated lipids, a moderately high Gothenburg (PPSW, Bengtsson et al. 1997; Lissner intake of fish (depending on the proximity of the sea), et al. 2003). Response rate decreased from 84% and a low-to-moderate intake of dairy products (and then 86% for men and women, respectively, in the earliest mostly in the form of cheese and yogurt), a low intake birth cohort to 65% for both sexes in the most recent of meat and meat products as well as a regular but cohort. Overall, a total of 1,277 men and women still moderate intake of ethanol, primarily in the form of residing in the Gothenburg district had undergone diet wine and generally during meals (Willett et al. 1995). history examinations at the age of 70 (Eiben et al. The Mediterranean diet was first considered a 2004, 2005). Diet history was validated by comparison dietary pattern that protected against coronary heart of energy intake (EI) with estimated total energy disease (de Lorgeril et al. 1999). This was confirmed expenditure (TEE) by heart rate monitoring, activity very recently by a longitudinal study showing that diary, and doubly-labeled water (Rothenberg et al. Mediterranean diet is associated with a higher 1998). preservation of ventricular function and a more In order to exclude implausible dietary intakes, favorable prognosis after an acute coronary event subjects in the highest and lowest 5 percentiles of the (Chrysohoou et al. 2010). However, in previous ratio between energy intake and calculated basal studies, other beneficial effects on total mortality metabolic rate (BMR) were not included in the have been discovered (Trichopoulou et al. 2005). present analyses (N=149). BMR was predicted from Interestingly, a recent paper on the Swedish popula- weight (W) by means of standard equations for subjects aged 60–74 (males: 0.0499×W+2.93; females: tion showed a reduction in total mortality among young women (Lagiou et al. 2006). In addition, a 0.0386×W+2.875; Department of Health 1991). Other Mediterranean dietary pattern has also been shown to exclusions included cases missing information on diet increase longevity among European elderly of the or on potential confounders (N=91). The final study Healthy Aging: a Longitudinal study in Europe sample included 1,037 subjects (52.1% females). All (HALE) project (Knoops et al. 2004). cohorts have been monitored continuously for mortal- In the present study, we investigated the associa- ity by linking personal identification numbers with the tion of the Mediterranean diet pattern with total national death registration system. The most recent mortality in a population of Swedish 70 year olds. mortality follow-up was in 2009. In a sub-sample of 781 subjects, a follow up at 75 years for both weight and waist circumference was available. In supplemen- Subjects and methods tary analyses, observation time was truncated to 8.5 years to give all cohorts equal follow-up time Subjects, dietary assessment, and outcome definition when estimating survival probabilities. This study is based on the Gerontological and Definitions of food groups and of the Mediterranean Geriatric Population Studies in Gothenburg (H70; diet scores Eiben et al. 2004, 2005; Rothenberg et al. 1996, 1997, 1998). The Revenue Office of Gothenburg has a To assess the association of diet and lifestyle factors registry of all inhabitants, from which representative with total mortality, a refined version of the modified study groups of specific ages may be identified. Using Mediterranean diet score (refined mMDS) was calcu- this registry, population-based samples of 70-year-old lated, based on existing knowledge from the scientific residents were recruited for health examinations in literature on positive effects of wholegrain cereals 1971, 1981, 1992, and 2000. In the first two (Flight and Clifton 2006) and moderate alcohol intake examinations, 70-year-old cohorts born in 1901 and (Mukamal et al. 2010) as well as on the fact that 1911 were sampled and invited to participate based on polyunsaturated fats (PUFA) and not only monoun- date of birth. In the latter two examinations, men born saturated fats (MUFA) are the principal unsaturated in 1922 and 1930 were sampled in the same way, fats in non-Mediterranean diets (De Lorgeril et al. AGE (2011) 33:439–450 441 1994). The score comprised nine components: (1) the outcome of the iso-energetic replacement of 5% of vegetables and potatoes, (2) legumes, nuts, and seeds, energy provided by the reference macronutrient (the (3) fruit and fresh juices, (4) wholegrain cereals, (5) one not included in the model, e.g., carbohydrates) fish and fish products, (6) ratio of MUFA+PUFA to with a similar amount of calories provided by all the saturated fats (SAFA), (7) alcohol intake, (8) meat, other macronutrients included in the model (e.g., meat products, and eggs, (9) dairy products. Intake of proteins and fats). In these analyses, alcohol intake each component was adjusted to daily energy intakes was added separately on a g/day basis. of 2,500 kcal (10.5 MJ) for men and 2,000 kcal Next, analyses were performed both on the Medi- (8.5 MJ) for women. Although these values are not terranean diet scores (HALE and refined mMDS) on a considered as recommended daily energy intake for continuous scale as well as on scores of 6–9 in refined both genders, they were chosen based on previous mMDS or 5–8 in HALE mMDS in comparison with research on this subject (Knoops et al. 2004) and used the others. In order to examine if the estimates for to obtain energy adjusted associations in all the mortality were influenced by any of the covariates, or if analyses. The sex-specific median intakes were taken any of these could remove part of the effect, the crude as cut-off points. In any case, the final score is hazard ratios (HRs) for the dietary and lifestyle factors considered a measure of relative adherence to the were also calculated and compared with the adjusted Mediterranean diet pattern. The diet score varied from HRs. Finally, the mortality risk was also estimated in 0 (low-quality diet) to 9 (high quality diet). For each single food groups contributing to these scores. Food of the nine components, except for meat and dairy intakes were re-scaled according to medians, as done for products, a value of 1 was assigned to subjects whose the score calculations. For illustration, the dose– consumption was higher than the sex-specific median, response effect of the mortality risk in subjects with growing levels of refined mMDS compared to the risk and 0 to the others. For meat and dairy products, the reverse rule was applied. An alternative version of this in subjects characterized by a score ≤1 was investigated. score, previously used in the HALE project (HALE In addition, a series of sensitivity analyses were mMDS, Knoops et al. 2004) was also produced, performed, the most critical of which was the estima- including only MUFA in the numerator of the fat tion of the association between mortality risk and ratio, total instead of wholegrain cereals, and exclud- different versions of the refined mMDS, calculated by ing alcohol intake from the score. excluding, one at a time, each component of the score in order to determine that the effect was not driven by a Statistical analyses single food group. Other sensitivity analyses included adjusting for cardiovascular risk factors such as weight Cox proportional hazards models were used to and waist circumference change, baseline blood pres- estimate the effects of diet, smoking status, education, sure, fasting glucose, total cholesterol, and triglycerides physical activity, and other potential confounders on or for activities of daily living variable (ADL, based on total mortality, to November 2009. All the models used need of help for everyday activities). The activities of in this study were adjusted for potential confounders daily living variable was based on the Katz index, such as sex, baseline body mass index (BMI), waist which is similar to the Barthel Index, and hence based circumference, smoking status, physical activity level, on the need of help for at least one out of nine different marital status, education, and birth cohort. activities, including house cleaning, shopping, transport, Analyses of macronutrients and mortality were first cooking, washing, dressing, using the toilet, ambulation, performed by means of the standard multivariate model and feeding (Sonn and Asberg 1991). Furthermore, (Willett 1998), in which each macronutrient was tested analyses were repeated after excluding the first 2 years separately, with or without adjustment for energy of total mortality follow-up. intake. Further analyses were conducted by means of We also repeated the Cox analysis on the refined the multivariate nutrient density model (Willett 1998), mMDS, by censoring at a fixed follow-up duration of in which the nutrient densities (percentages of energy 8.5 years (approximately the longest follow-up time from non-alcohol sources in a scale of 5% units) from in the latest-born cohort) instead of censoring at a total fat and protein intakes are included as indepen- fixed date. Crude proportions of subjects still alive at dent variables. The model thus estimates the effect on 8.5 years were also calculated according to three 442 AGE (2011) 33:439–450 different levels of the score (≤3, 4–6, ≥7) and the consistently high (1.5±0.3), indicating a good quality association between adherence to the Mediterranean of nutritional data. In the subjects excluded for diet pattern and survival was assessed by means of a implausible dietary intakes (highest and lowest 5 chi-square test. percentiles of energy intake vs. basal metabolic rate Finally, potential confounders (see above) were ratio), EI/BMR was either less than 0.9 or greater than examined in more detail, including an estimation of 2.2 (data not shown). the mortality HRs in subjects characterized by a score Regarding the two scores measuring the adherence to in the highest four levels (>4 for the HALE mMDS, the Mediterranean diet (Table 2), no major differences >5 for the refined mMDS) compared to the other appeared from cohort to cohort, although for the subjects. Moreover, the assessment of the effect of all refined mMDS a slight but significant increase was main covariates on mortality included in the previous evident (4.3±1.6 in 1901 cohort to 4.8±1.8 in 1930 models was also performed. cohort), probably due to the increase in wholegrain, All statistical analyses were considered significant alcohol, and PUFA intakes included in this score and under a p value of 0.05 and were carried out using the not in the HALE mMDS. SAS statistical software version 9.2 (SAS Institute, Table 3 shows the descriptive analyses of potential Cary, NC, USA). confounders. Both genders were equally represented in the population (47.9% men). Subjects characterized Bioethics by a BMI <20 were only a minority (3.1%), while obese subjects constituted 15.8% of the total popula- The latest H70 examinations (since 1992) have tion. A high waist circumference (>88 cm in women, been approved by the Gothenburg University Ethics >102 cm in men), characterized 31.0% of the subjects. Only 14.2% of subjects were physically Committee in accordance with the Declaration of Helsinki (1989) of the World Medical Association. inactive, 62.1% were married, and more than 30% of All participants were informed of the aims and subjects were smokers (or stopped less than 10 years procedures of the study and gave their consent. ago) or had an education above basic (i.e., 6 years of schooling or more). Overall, in the studied popula- tion, there were 630 deaths (60.8%) until November Results 2009: 309 women and 321 men. Regarding main cardiovascular risk factors depicted Descriptive analyses in Table 3, which include variations in anthropometry over 5 years, we observed a decrease in mean weight Main descriptive analyses are reported in Tables 1 from 74.2 kg±12.9 at baseline to 72.3 kg±13.2 at (dietary variables), 2 (Mediterranean diet scores), and 3 follow up (mean change: −1.9 kg±5.4), a quite stable (potential confounders and other cohort descriptors). waist circumference (mean change: −0.2 cm±7.4) and Dietary intake distributions across the birth cohorts BMI (mean change: −0.3 kg/m ±1.9). Diastolic blood have previously been described (Eiben et al. 2004; pressure means were less than 90 mmHg (87.4 mmHg± Rothenberg et al. 1996). Table 1 shows that, according 11.9), while systolic blood pressure mean was above to trend test, across the four birth cohorts there was an 140 mmHg (159.0 mmHg±23.6). Regarding plasma increase in the consumption of vegetables and potatoes, parameters (only available at baseline), fasting fruit, legumes plus nuts and seeds, fish and fish glucose was equal to 100.8 mg/dL±36.0, total choles- products, meat and meat products, red wine and terol was 233.3 mg/dL±77.8 while triglycerides was alcohol. In contrast, a decrease in the consumption of 132.4 mg/dL±70.6. cereals (both total and wholegrain) was observed. No clear patterns emerged for dairy products, MUFA/ Diet quality and mortality SAFA ratio and (MUFA+PUFA)/SAFA ratio. Regarding macronutrient intakes, a trend in protein, carbohydrate, In the present cohort, the total mortality was around and fat intake was observed. Energy intakes showed a 60%, with a higher rate in the earlier born cohorts and slight tendency to an increase across the cohorts, and a lower in the later born cohorts (100% in 1901, 98% the ratio between energy intake and BMR was in 1911, 48% in 1922, and 15% in 1930 birth cohort). AGE (2011) 33:439–450 443 Table 1 Means and standard deviation in comparison with medians and 95% Confidence Limits (CLs) of dietary variables used in the analyses, including test for trend across the birth cohorts e,f Food groups/macronutrients Overall mean (N=1,037) Median intakes and 95% CLs p for trend across birth cohorts a,b Vegetables and potatoes (g/day) 237.6±98.7 ♀: 209.5 (99.6; 406.3) <0.0001 ♂: 239.0 (120.1; 432.2) a,b Fruit (g/day) 196.6±146.3 ♀: 176.4 (22.3; 527.7) <0.0001 ♂: 155.5 (14.0; 456.0) a,b Legumes nuts and seeds (g/day) 15.2±20.2 ♀: 2.0 (0; 40.0) <0.0001 ♂: 13.3 (0; 60.0) Cereals (g/day) 207.4±104.0 ♀: 165.0 (68.2; 383.0) (−)<0.0001 ♂: 213.0 (98.3; 442.1) Wholegrain cereals (g/day) 107.9±95.6 ♀: 74.2 (0; 298.5) (−)<0.0001 ♂: 92.8 (0; 322.0) a,b Fish and fish products (g/day) 53.8±35.8 ♀: 45.2 (12.8; 105.3) <0.0001 ♂: 53.7 (13.8; 129.5) a,b Dairy products (g/day) 445.1±251.7 ♀: 373.3 (127.9; 829.5) 0.41 ♂: 446.0 (74.0; 1061.9) Meat and meat products (g/day) 105.4±47.9 ♀: 89.7 (38.4; 168.8) <0.0001 ♂: 109.1 (52.5; 204.9) Meat, meat products, eggs (g/day) 129.5±55.3 ♀: 110.1 (47.9; 187.3) <0.001 ♂: 137.7 (66.3; 251.1) Red wine (g/day) 2.0±4.7 ♀: 0 (0; 9.8) <0.001 ♂: 0 (0; 10.7) Alcohol (g/day) 6.0±9.1 ♀: 1.3 (0; 13.9) <0.0001 ♂: 5.3 (0; 28.5) Carbohydrate (g/day) 249.1±64.8 ♀: 217.1 (143.5; 316.8) 0.01 ♂: 272.4 (192.8; 398.2) Protein (g/day) 80.1±20.0 ♀: 71.2 (48.6; 100.4) <0.0001 ♂: 85.2 (58.3; 129.0) Fat (g/day) 86.1±25.4 ♀: 74.6 (46.6; 110.) <0.0001 ♂: 93.5 (60.1; 146.5) MUFA/SAFA 0.8±0.2 ♀: 0.8 (0.6; 1.1) 0.40 ♂: 0.8 (0.6; 1.1) (MUFA+PUFA)/SAFA 1.2±0.3 ♀: 1.1 (0.8; 1.6) 0.50 ♂: 1.1 (0.7; 1.7) Energy intake (100 kcal) 21.4±5.0 ♀: 18.5 (13.3; 25.4) 0.04 ♂: 23.6 (17.7; 33.6) Energy/BMR 1.5±0.3 ♀: 1.4 (1.0; 1.9) 0.30 ♂: 1.4 (1.1; 2.1) Included in HALE mMDS Included in refined mMDS Included in place of alcohol in an additional analysis Not included in any scores, just included for descriptive purposes Birth–cohort effect from a regression model adjusted for gender, BMI, waist circumference, physical activity, smoking status, marital status and education All significant trends positive in direction except if indicated with (−) In the first part of the present work, we analyzed the intake (adjusted or unadjusted for energy intake) in association between diet macronutrient composition standard multivariate models. Alcohol was also not and total mortality. No clear association emerged associated on a continuous scale with total survival either when analyzing protein, fat, and carbohydrate time (Table 4). Similar results were obtained when 444 AGE (2011) 33:439–450 Table 2 General descriptive of Mediterranean diet score distributions, both stratified by birth cohort and for the overall population Mediterranean diet score 1901 (N=323) 1911 (N=214) 1922 (N=88) 1930 (N=412) Overall (N=1,037) p for trend across birth cohorts Refined mMDS (mean±SD) 4.3±1.6 4.0±1.6 4.8±1.5 4.8±1.8 4.5±1.7 <0.001 Medians (5th;95th perc) 4 (2; 7) 4 (1; 7) 5 (2; 7) 5 (2; 8) 4 (2; 7) HALE mMDS (mean±SD) 4.1±1.5 3.6±1.4 4.1±1.4 4.1±1.6 4.0±1.5 0.02 Medians (5th;95th perc) 4 (2; 6) 4 (1; 6) 4 (2; 6) 4 (1; 7) 4 (2; 6) Birth–cohort effect from a regression model adjusted for gender, BMI, waist circumference, physical activity, smoking status, marital status, and education employing the nutrient density model to describe the In light of the null results obtained by the analyses effect of the reciprocal substitution of 5% of energy of macronutrients, we then focused on the identifica- intake from each macronutrient with carbohydrates tion of a possible dietary pattern, which could be (Table 4). related to total mortality. In particular, we chose to study the Mediterranean diet pattern by means of a Table 3 Main covariate frequencies and cohort description refined version of the modified Mediterranean Diet Potential confounders Frequency (%) Score (refined mMDS), which we consider to reflect a closer adherence to the classic Mediterranean diet Total sample (N = 1,037) pattern. Male gender 47.9 Table 5 shows the association of refined mMDS, in BMI<20 3.1 comparison to the previously used HALE mMDS BMI>30 15.8 (Knoops et al. 2004) with mortality risk, as well as the High waist circumference 31.0 same outcome for those food groups on whose intakes Low physical activity at 70 14.2 the adherence to Mediterranean diet pattern was Married at age 70 62.1 assessed. Smoker/stopped <10 years ago 30.6 Regarding Mediterranean diet score, an inverse School education above basic 30.5 association with total mortality was shown for the Activities of daily living (ADL) 12.7 continuous refined mMDS (HR=0.93, 95% CI: 0.89; 0.98), while no significant association emerged with Additional covariates Means±SD HALE mMDS, although the trend was toward an Baseline diastolic blood pressure (mmHg) 87.4±11.9 inverse association (HR=0.97, 95% CI: 0.92; 1.02). Baseline systolic blood pressure (mmHg) 159.0±23.6 Moreover, the comparison of the lowest-risk group Baseline fasting glucose (mg/dL) 100.8±36.0 Baseline total cholesterol (mg/dL) 233.3±77.8 Table 4 Hazard ratios and 95% confidence limits from a Cox- Baseline triglycerides (mg/dL) 132.4±70.6 proportional hazard model (adjusted for gender, baseline BMI, waist circumference, physical activity, marital status, smoking status, birth cohort, and education) estimating the association of Subsample with follow up at 75 years (N = 781) macronutrient with mortality risk and different substitution of Changes in anthropometry Means±SD energy from each macronutrient with the same amount of energy of another one Weight change (kg) −1.9±5.4 Baseline 74.2±12.9 Macronutrient (10 g) HR 95% CLs Follow up 72.3±13.2 Waist circ. change (cm) −0.2±7.4 Alcohol 0.98 0.86; 1.11 Baseline 90.0±11.4 Protein 0.97 0.90; 1.04 Follow up 90.0±11.8 CHO 1.00 0.98; 1.03 BMI change (kg/height ) −0.3±1.9 Fat 1.00 0.94; 1.07 Baseline 26.4±3.9 From CHO to protein 5% energy 0.95 0.80; 1.14 Follow up 26.1±4.1 From CHO to fat 5% energy 1.01 0.94; 1.09 AGE (2011) 33:439–450 445 Table 5 Hazard ratios and 95% confidence limits from a Cox- 0.71; 0.98). In contrast, other food groups were not proportional hazard model (adjusted for gender, baseline BMI, significantly associated with mortality, such as vege- waist circumference, physical activity, marital status, smoking tables and potatoes (HR=1.06, 95% CI: 0.90; 1.24), status, birth cohort, and education) estimating the association of both HALE and refined mMDS with mortality risk fruit (HR=1.03, 95% CI: 0.87; 1.21), legumes plus nuts and seeds (HR=0.98, 95% CI: 0.83; 1.16), and Mediterranean diet score or food group HR 95% CLs fish (HR=0.96, 95% CI: 0.82; 1.13). No association was shown for either of the fat ratios (MUFA/SAFA Refined mMDS 0.93 0.89; 0.98 ratio: HR=0.98, 95% CI: 0.84: 1.15; (MUFA+PUFA)/ Crude estimate 0.92 0.88; 0.97 SAFA ratio: HR=0.96, 95% CI: 0.82; 1.13). It is worth Highest 4 levels vs. the others 0.82 0.67; 0.99 mentioning that vegetables alone and potatoes alone Crude estimate 0.81 0.67; 0.99 did not show any association with total mortality (data HALE mMDS 0.97 0.92; 1.02 not shown). Crude estimate 0.97 0.92; 1.03 In Fig. 1 dose–response analysis results are Highest 4 levels vs. the others 0.94 0.79; 1.11 depicted. Briefly, the group characterized by a refined Crude estimate 0.94 0.80; 1.11 mMDS equal to 0 or 1 was the reference, compared to High intake/level of: the other groups with increasing levels of the score. Vegetables and potatoes 1.06 0.90; 1.24 Although none of the single comparisons reached Fruit 1.03 0.87; 1.21 statistical significance, the analysis showed a dose– Legumes, nuts, and seeds 0.98 0.83; 1.16 response tendency, thus suggesting a decrease in the Cereals 1.01 0.86; 1.19 mortality risk as long as the adherence to the Wholegrain cereals 0.85 0.73; 1.00 Mediterranean diet pattern increased. Fish 0.96 0.82; 1.13 Alcohol 0.77 0.61; 0.97 Supplementary analyses MUFA/SAFA ratio 0.98 0.84; 1.15 (MUFA+PUFA)/SAFA ratio 0.96 0.82; 1.13 A sensitivity analysis was performed on the refined Low intake of: mMDS, with the aim of understanding whether any Dairy products 0.82 0.70; 0.96 one of the food group items included in the score Meat and meat products 0.89 0.76; 1.05 could invalidate the effect of the entire score, with Meat, meat products and eggs 0.84 0.71; 0.98 respect to mortality. Nine different scores were then produced, each excluding one item at a time (and keeping all others). Table 6 clearly demonstrates that (highest four levels of the score) versus the other the association of the refined mMDS with mortality is robust and survives all the item-by-item exclusions. subjects showed a significant inverse association for the refined mMDS (HR=0.82, 95% CI: 0.67; 0.99), but Furthermore, in order to test whether the positive not for the highest levels of the HALE mMDS (HR= effect of alcohol intake within the score could be 0.94, 95% CI: 0.79; 1.11; Table 5). Crude estimates obtained considering red wine (instead of total alcohol were also calculated and, as shown in the table, they are intake from all alcoholic beverages), considered a more quite overlapping with the adjusted ones, thus showing accurate indicator of the Mediterranean alcoholic that covariates did not have any strong influence on drinks, we ran an additional analysis, ending up with results. The protective effect of the Mediterranean diet a similar outcome and a slightly stronger hazard ratio pattern was stronger in the two youngest birth cohorts (HR: 0.92, 95% CI: 0.87; 0.97; Table 6). compared to the others (data not shown). In order to confirm that the effect of diet on mortality When studying the effect of single food groups, an was independent of the subjects’ functional capacity levels, and thus that the effect of Mediterranean diet inverse association with total mortality was shown for high intakes of wholegrain cereals (HR=0.85, 95% was not entirely explained by how impaired the CI: 0.73; 1.00), alcohol consumption (HR=0.77, 95% subjects were at the time of diet assessment we also CI: 0.61; 0.97) and low intake of dairy products (HR= repeated all the models adjusting for activities of daily 0.82, 95% CI: 0.70; 0.96) as well as low intakes of living variable and found no substantial change in meat, meat products, and eggs (HR=0.84, 95% CI: results. Similar analyses were performed adjusting for 446 AGE (2011) 33:439–450 Fig. 1 Dose–response analyses based on Cox- proportional hazard model (adjusted for gender, baseline BMI, waist circumference, physical activity, marital status, smoking status, birth cohort and education) comparing mortality risk in subjects characterized by a refined mMDS ≤1 with the same risk in subjects in the other score categories. Dots and vertical lines indicate HR and 95% confidence limits other cardiovascular risk factors, i.e., baseline blood more, to minimize the possibility that diet or lifestyle pressure, fasting glucose, total cholesterol, and trigly- factors had changed in response to subclinical diseases, cerides; this did not strongly affect the results, except all analyses were also repeated after exclusion of the that the association of the dichotomous score (higher vs. subjects who had died in the first 2 years of follow up. lower) was attenuated (HR=0.85, 95% CI: 0.70; 1.04). Here again, no clear difference emerged, thus confirm- Adjustment for weight or waist circumference change ing our main results. did not substantially affect the association of the refined Finally, the association of the refined mMDS with mMDS with mortality. However, both were inversely total mortality was also tested in a Cox model in associated with total mortality (weight change: HR= which subject censoring was done at a fixed follow up 0.95, 95% CI: 0.93; 0.97; waist circumference change: time (8.5 years, the maximum in the youngest birth HR=0.98, 95% CI: 0.97; 1.00), suggesting a greater cohort) instead at a fixed date. The previous results mortality risk in association with weight loss. Further- were confirmed although somewhat attenuated (HR= 0.92, 95% CI: 0.86; 1.00). The distribution of crude proportions of survival rates at 8.5 years, stratified by increasing levels of the refined mMDS were as Table 6 Hazard ratios and 95% confidence limits from a Cox- proportional hazard model-based sensitivity analysis estimating follows: 67% survival among those characterized by the effect of the exclusion of each level of the refined mMDS+ a score ≤3,77% survival at intermediate levels of the alcohol on mortality risk (adjusted for gender, baseline BMI, score (from 4 to 6) and 80% survival among the waist circumference, physical activity, marital status, smoking subjects with a score ≥7(p=0.02). status, birth cohort, and education) Mediterranean diet score HR 95% CLs Effect modification and covariates associated with diet and mortality Refined mMDS 0.93 0.89; 0.98 Without fat ratio 0.92 0.87; 0.97 In order to identify potential effect modifiers, statis- Without vegetables and potatoes 0.91 0.87; 0.96 tical interactions between all dietary and potential Without fruit 0.92 0.87; 0.97 confounders were tested, but none of them were Without legumes, nuts and seeds 0.93 0.88; 0.97 statistically significant at a 0.05 level. The association Without cereals 0.94 0.89; 0.99 of the highest levels of the refined mMDS with each Without fish 0.92 0.87; 0.97 of the above-mentioned confounders was also inves- Without dairy 0.94 0.89; 0.99 tigated (Table 7). Subjects ranked to the highest levels Without meat 0.94 0.90; 0.99 of the score were less likely to be inactive (OR=0.42; Without alcohol 0.94 0.89; 0.99 95% CI: 0.25; 0.71) and more likely to be married Refined mMDS (including red wine 0.92 0.87; 0.97 (OR=1.47, 95% CI: 1.07; 2.01), more educated instead of alcohol) (OR=1.38, 95% CI: 1.02; 1.86) or born within the AGE (2011) 33:439–450 447 Table 7 Hazard ratios and Confounders Association with highest Association with 95% confidence limits from levels of refined mMDS mortality a Cox-proportional hazard model estimating the OR 95% CLs HR 95% CLs association of potential confounders included in the Male gender 0.90 0.66; 1.23 2.02 1.67; 2.44 final model and mortality risk in the whole studied BMI<20 1.06 0.45; 2.50 1.32 0.90; 1.96 population (N=1,037) BMI>30 0.92 0.56; 1.51 0.94 0.72; 1.24 High waist circumference 0.80 0.54; 1.18 1.29 1.04; 1.60 Low physical activity at 70 0.42 0.25; 0.71 1.29 1.04; 1.59 Married at age 70 1.47 1.07; 2.01 0.78 0.66; 0.93 Smoker/stopped <10 years ago 1.07 0.77; 1.47 1.37 1.14; 1.64 School education above basic 1.38 1.02; 1.86 0.91 0.75; 1.11 Birth cohort 1.03 1.01; 1.04 0.97 0.97; 0.98 more recent birth cohorts (OR 0 1.03, 95% CI: the time of the dietary assessment, or biased in some 1.01; 1.04). way by how impaired they were at enrolment. Table 7 also shows the association of the potential Furthermore, the effect was only slightly attenuated confounders included in the final model with total when adjusting for several indicators of cardiovascu- mortality. As expected, there is a positive association lar risk, suggesting that the protective effect was not between the risk of mortality and male gender (HR= limited to high-risk subjects. 2.02, 95% CI: 1.67; 2.44), high waist circumference Other studies had already found a positive associ- (HR=1.29, 95% CI: 1.04; 1.60), low level of physical ation of Mediterranean diet and health (Lagiou et al. activity (HR=1.29, 95% CI: 1.04; 1.59) and smoking 2006; Knoops et al. 2004; Trichopoulou et al. 1995), (HR=1.37, 95% CI: 1.14; 1.64). Inverse associations showing the effects of increasing the intake of were found for marital status (HR=0.78, 95% CI: vegetables and fruit, fish, and cereals, while decreas- 0.66; 0.93), and for birth cohort, which we considered ing animal products such as dairy and meat. In our as a proxy for age/period effects (HR=0.98, 95% CI: study, we started by reproducing the same mMDS that 0.97; 0.98). Finally, BMI did not show any significant was previously created by the HALE working group association. for their paper on Mediterranean diet and total mortality (Knoops et al. 2004) and did not find a significant trend, even when testing the association at the highest Discussion levels. However, we found an inverse association with total mortality when a refined version of the mMDS In this paper, we investigated the association of a was tested instead of the original score. The associa- dietary pattern close to the Mediterranean diet with tion was indeed strong and robust after calculating the total mortality in population-based cohorts of 70-year- score considering wholegrain cereals instead of total old Swedes. Although we did not find an association cereals, adding alcohol (or red wine) intake, consider- when testing the previously defined HALE mMDS on ing egg intake together with meat products and our Swedish population (Knoops et al. 2004), we calculating the fat ratio including PUFA and not only were able to refine this score in order to obtain one MUFA in the numerator. The addition of alcohol that best described strict adherence to a Mediterranean intake should not be considered problematic, since the intakes in this population were not high (75th diet-like pattern. None of the items included in the score were found to be essential for the association percentile: 7.6 g/day, 90th percentile: 16.3 g/day) and with the total score. Finally, the association was not can then be considered in an order of magnitude quite attenuated when adjusting for ADL, thus showing that close to international recommendations. Interestingly, the protection by the Mediterranean pattern was not beneficial effects of moderate alcohol intake have limited to those subjects still living independently at been recently confirmed in a study with a follow up of 448 AGE (2011) 33:439–450 20 years that also considered many confounders estimated intakes with National Surveys; Becker typically associated with abstaining (such as past 1994), the possibility of adjusting for different known history of heavy drinking behavior, Holahan et al. confounders and the availability of weight and waist 2010). Furthermore, our association was also con- circumference measurements both at baseline and at firmed when including red wine instead of alcohol, follow up. The limitations include the lack of repeated confirming that the positive result characterizes the dietary assessment and the small size of the study subjects adhering to the healthiest diet patterns. population. Moreover, it must be acknowledged that The possible explanation of the necessity of improv- diet patterns alone may inherently co-vary with other ing an existing score that had already worked in health-related phenomena (e.g., healthy lifestyles, previous studies to find a robust association in the illnesses, weight status). Thus, it is important to note Swedish elderly may be due to the obvious differences that associations were independent of cardiovascular between the original Mediterranean diet and the risk factors, weight and waist circumference change, Swedish Mediterranean-like diet. This is particularly SES, education, ADL, and other risk factors. More- important in our population, since many subjects were over, it is worth mentioning that although the analyses born at the beginning of last century, when many were always adjusted for birth cohort, the results products that were common in Southern Europe were could still be influenced by a residual cohort effect as not consumed in large quantities in the north. Swedish suggested by the fact that the protective effect of the food habits have indeed undergone major changes in the Mediterranean diet pattern was stronger in the last 30 years, particularly in relation to consumptions of youngest cohorts. This was probably due to a higher fresh fruit and vegetables (now available in quantity quality of nutritional data (or higher health conscious- even on the Scandinavian market) and cereals as well as ness in the subjects) in the most recent surveys. From the above mentioned results, it clearly a higher unsaturated fat proportion (Eiben et al. 2004). Another reasonable example is represented by MUFA emerges that it would be useful to put more emphasis sources, mainly olive oil in Southern European on dietary recommendations directed to the elderly in countries, mostly margarines (a source of trans fatty order to encourage increased consumption of fruit and acids, especially at the time they were measured) in the vegetables, wholegrain cereals and fish, while reduc- oldest birth cohorts and only marginally derived from ing the intake of dairy and meat products, in favor of olive oil in the 1930 birth cohort. other healthier protein sources such as legumes. This However, by applying stricter rules, such as is particularly important in Sweden, considering that substituting wholegrain cereals instead of total cereals current guidelines do not always place sufficient (often characterized by a higher sugar content than in emphasis on this type of recommendation. For South Europe), including PUFA intake (in Sweden, instance, although our results are quite in line with fish is an important contributor), adding alcohol (and Nordic Nutrition Recommendations (Nordic Council also red wine only), and including egg intake (a of Ministers 2004), the present Swedish Nutrition probable indicator of a low-quality diet), we found a Recommendations Objectified (Barbieri and Lindvall protective pattern in our Swedish population. 2005), is in some ways different from Mediterranean Another important issue is that single food groups diet-based features. Indeed, the former considers were not always found to be inversely or directly acceptable a daily intake of 500 ml milk (in addition associated with the mortality risk, although the full to other dairy products) and suggests some equilibrated score is robust to exclusions of any single component. menus that do not consider olive oil either as a This suggests that the use of this and similar scores in condiment or as a cooking fat, while including nutritional epidemiology studies is indeed a useful margarine on a daily basis. Moreover, it often refers to strategy for investigating associations between diet “refined rye bread” and “white bread” instead of and health outcomes. suggesting wholegrain products, and includes meat Our study has both strengths and limitations, the products in two meals per day. At the same time, it is former being the high quality of nutritional data also worth mentioning that carbohydrate restriction is obtained by a validated diet history during a face to becoming an increasingly popular (although not scien- face interview with the dietician (confirmed by the tifically based) weight control method in the popula- high EI/BMR ratio and by concordance of our tion, a modification which is also not supported, vis-à- AGE (2011) 33:439–450 449 vis mortality, in our nutrient-level analyses simulating References carbohydrate replacement. Overall, studies of dietary patterns are inherently Appel LJ (2008) Dietary patterns and longevity. Circulation complex. However, regardless of scientific approach, 118:214–215 there is a remarkable convergence of evidence on the Barbieri HE, Lindvall C (2005) De svenska näringsrekommen- dationerna översatta till livsmedel: underlag till generella fact that dietary patterns associated with longevity råd på livsmedels-och måltidsnivå för friska vuxna emphasize fruits and vegetables and are reduced in Swedish Nutrition Recommendations Objectified (SNO). saturated fat, meats, refined grains, sweets, and full- Basis for general advice on food consumption for healthy fat dairy products (Appel 2008). This is not only the adults. Livsmedelsverkets rapport n. 20/2005. http://www. slv.se/upload/dokument/rapporter/mat_naring/ case for the Mediterranean diet pattern, but also for Report_20_2005_SNO_eng.pdf other pattern such as the DASH diet (Parikh et al. Becker BW (1994) Dietary habits and nutrient intake in 2009) or the Okinawa diet (Willcox et al. 2009). Sweden 1989. Statens livsmedelsverk, Uppsala, 240 pp Bengtsson C, Ahlqvist M, Andersson K, Björkelund C, Lissner L, Söderström M (1997) The prospective population study of women in Gothenburg, Sweden, 1968–69 to 1992–93. Conclusions A 24-year follow-up study with special reference to participation, representativeness, and mortality. Scand J To conclude, we can reasonably state that the Prim Health Care 15:214–219 Chrysohoou C, Panagiotakos DB, Aggelopoulos P, Kastorini adherence to a Mediterranean-like dietary pattern is CM, Kehagia I, Pitsavos C, Stefanadis C (2010) The inversely related to total mortality also in a Swedish Mediterranean diet contributes to the preservation of left population of elderly subjects. Our hope is that the ventricular systolic function and to the long-term favorable results of the present research will stimulate a prognosis of patients who have had an acute coronary event. Am J Clin Nutr 92:47–54 productive discussion on these issues and be consid- De Lorgeril M, Renaud S, Mamelle N, Salen P, Martin JL, ered in updated food- and nutrient-level guidelines for Monjaud I, Guidollet J, Touboul P, Delaye J (1994) the population. Mediterranean alpha-linolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet 343:1454–1459 de Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Acknowledgment The authors’ responsibilities were as fol- Mamelle N (1999) Mediterranean diet, traditional risk lows—GT performed the data analysis and wrote the manu- factors, and the rate of cardiovascular complications after script; LL coordinated the research, contributed to the myocardial infarction: final report of the Lyon Diet Heart interpretation of results and to the writing of the manuscript; Study. Circulation 99:779–785 ER and GE provided support for the correct application of Department of Health (1991) Report on health and social nutritional data, helped with the interpretation of results and subjects; 41. Dietary reference values for food energy and gave critical comments on the manuscript; AW contributed to nutrients for the United Kingdom. HMSO, London the interpretation of results and gave critical comments on the Eiben G, Andersson CS, Rothenberg E, Sundh V, Steen B, manuscript; VS provided statistical expertise and gave critical Lissner L (2004) Secular trends in diet among elderly comments on the manuscript. The research was funded by the Swedes—cohort comparisons over three decades. Public Swedish Council on Working Life and Social Research (FAS) Health Nutr 7:637–644 EpiLife centre. Eiben G, Dey DK, Rothenberg E, Steen B, Björkelund C, Bengtsson C, Lissner L (2005) Obesity in 70-year-old Financial disclosures All authors have no financial disclosures. Swedes: secular changes over 30 years. Int J Obes (Lond) 29:810–817 Flight I, Clifton P (2006) Cereal grains and legumes in the Conflict of interest None declared. prevention of coronary heart disease and stroke: a review of the literature. Eur J Clin Nutr 60:1145–1159 Haveman-Nies A, de Groot L, van Staveren W (2003) Relation Disclaimer The views expressed are those of the authors and of dietary quality, physical activity, and smoking habits to should not be construed to represent the positions of anybody 10-year changes in health status in older Europeans in the else. SENECA study. Am J Public Health 93:318–323 Huijbregts P, Feskens E, Rasanen L, Fidanza F, Nissinen A, Menotti A, Kromhout D (1997) Dietary pattern and 20 year Open Access This article is distributed under the terms of the mortality in elderly men in Finland, Italy, and the Nether- Creative Commons Attribution Noncommercial License which lands: longitudinal cohort study. BMJ 315:13–17 permits any noncommercial use, distribution, and reproduction Keys AB (1980) Seven countries: a multivariate analysis of in any medium, provided the original author(s) and source are death and coronary heart diseases. Harvard University credited. Press, Cambridge 450 AGE (2011) 33:439–450 Knoops KT, de Groot LC, Kromhout D, Perrin AE, Moreiras- Rothenberg E, Bosaeus I, Steen B (1997) Evaluation of energy Varela O, Menotti A, van Staveren WA (2004) Mediterra- intake estimated by a diet history in three free-living nean diet, lifestyle factors, and 10-year mortality in elderly 70 year old populations in Gothenburg, Sweden. Eur J European men and women: the HALE project. JAMA Clin Nutr 51:60–66 292:1433–1439 Rothenberg E, Bosaeus I, Lernfelt B, Landahl S, Steen B Holahan CJ, Schutte KK, Brennan PL, Holahan CK, Moos BS, (1998) Energy intake and expenditure: validation of a diet Moos RH (2010) Late-life alcohol consumption and 20-year history by heart rate monitoring, activity diary and doubly mortality. Alcohol Clin Exp Res 34:1–11 labeled water. Eur J Clin Nutr 52:832–838 Lagiou P, Trichopoulou D, Sandin S, Lagiou A, Mucci L, Wolk Seymour J, Calle E, Flagg E, Coates R, Ford E, Thun M (2003) A, Weiderpass E, Adami HO (2006) Mediterranean dietary Diet quality index as a predictor of short-term mortality in pattern and mortality among young women: a cohort study the American Cancer Society Cancer Prevention Study II in Sweden. Brit J Nutr 96:384–392 Nutrition Cohort. Am J Epidem 157:980–988 Lissner L, Skoog I, Andersson K, Beckman N, Sundh V, Waern Sonn U, Asberg KH (1991) Assessment of activities of daily M, Zylberstein DE, Bengtsson C, Björkelund C (2003) living in the elderly. A study of a population of 76-year- Participation bias in longitudional studies—experiences olds in Gothenburg, Sweden. Scand J Rehabil Med from the The Population Study of women in Gothenburg. 23:193–202 Scand J Prim Health Care 21:242–247 Trichopoulou A, Kouris-Blazos A, Wahlqvist ML, Gnardellis Mai V, Kant AK, Flood A, Lacey JV Jr, Schairer C, Schatzkin C, Lagiou P, Polychronopoulos E, Vassilakou T, Lipworth A (2005) Diet quality and subsequent cancer incidence L, Trichopoulos D (1995) Diet and overall survival in and mortality in a prospective cohort of women. Int J Epid elderly people. BMJ 311:1457–1460 34:54–60 Trichopoulou A, Orfanos P, Norat T, Bueno-de-Mesquita B, Mukamal KJ, Chen CM, Rao SR, Breslow RA (2010) Alcohol Ocke MC, Peeters PH et al (2005) Modified Mediterra- consumption and cardiovascular mortality among US nean diet and survival: the EPIC-Elderly study. BMJ adults, 1987 to 2002. J Am Coll Cardiol 55:1328–1335 330:991 Nordic Council of Ministers (2004) Nordic Nutrition Recom- Willcox DC, Willcox BJ, Todoriki H, Suzuki M (2009) The mendations 2004. Integrating nutrition and physical Okinawan diet: health implications of a low-calorie, activity. Norden, Copenhagen nutrient-dense, antioxidant-rich dietary pattern low in Parikh A, Lipsitz SR, Natarajan S (2009) Association between glycemic load. J Am Coll Nutrition 28:500S–516S a DASH-like diet and mortality in adults with hyperten- Willett WC (1998) Nutritional epidemiology, 2nd edn. Oxford sion: findings from a population-based follow-up study. University Press, New York Am J Hypertens 22:409–416 Willett WC, Sacks F, Trichopoulou A, Drescher G, Ferro-Luzzi Rothenberg E, Bosaeus I, Steen B (1996) Food habits in three A, Helsing E, Trichopoulos D (1995) Mediterranean diet 70-year-old free-living populations in Gothenburg, Sweden. pyramid: a cultural model for healthy eating. Am J Clin A 22-year cohort study. Scand J Nutr 40:104–110 Nutr 61:1402S–1406S http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Age Pubmed Central

Does the Mediterranean diet predict longevity in the elderly? A Swedish perspective

Loading next page...
 
/lp/pubmed-central/does-the-mediterranean-diet-predict-longevity-in-the-elderly-a-swedish-raqvP6GJ6D

References (42)

Publisher
Pubmed Central
Copyright
© The Author(s) 2010
ISSN
0161-9152
eISSN
1574-4647
DOI
10.1007/s11357-010-9193-1
Publisher site
See Article on Publisher Site

Abstract

AGE (2011) 33:439–450 DOI 10.1007/s11357-010-9193-1 Does the Mediterranean diet predict longevity in the elderly? A Swedish perspective Gianluca Tognon & Elisabet Rothenberg & Gabriele Eiben & Valter Sundh & Anna Winkvist & Lauren Lissner Received: 14 July 2010 /Accepted: 28 October 2010 /Published online: 26 November 2010 The Author(s) 2010. This article is published with open access at Springerlink.com Abstract Dietary pattern analysis represents a useful censoring at fixed follow-up time, adjusting for improvement in the investigation of diet and health activities of daily living and main cardiovascular risk relationships. Particularly, the Mediterranean diet factors including weight/waist circumference changes pattern has been associated with reduced mortality at follow up. In conclusion, we can reasonably state risk in several studies involving both younger and that a higher adherence to a Mediterranean diet pattern, elderly population groups. In this research, relation- especially by consuming wholegrain cereals, foods ships between dietary macronutrient composition, as rich in polyunsaturated fatty acids, and a limited well as the Mediterranean diet, and total mortality amount of alcohol, predicts increased longevity in the were assessed in 1,037 seventy-year-old subjects (540 elderly. females) information. Diet macronutrient composition . . . was not associated with mortality, while a refined Keywords Elderly Mediterranean diet Diet quality . . version of the modified Mediterranean diet index Macronutrient intake Diet adherence Mortality showed a significant inverse association (HR=0.93, 95% CI: 0.89; 0.98). As expected, inactive subjects, smokers and those with a higher waist circumference Introduction had a higher mortality, while a reduced risk charac- terized married and more educated people. Sensitivity Epidemiologic studies indicate that quality of diet analyses (which confirmed our results) consisted of: together with other lifestyle factors have considerable exclusion of one food group at a time in the influence on health status as well as mortality risk Mediterranean diet index, exclusion of early deaths, (Huijbregts et al. 1997; Seymour et al. 2003; Haveman-Nies et al. 2003). Although many studies have examined effects of single nutrients, foods, or food : : : G. Tognon (*) G. Eiben V. Sundh L. Lissner groups on health status, it is becoming increasingly Public Health Epidemiology Unit, evident that the assessment of dietary patterns could be a Department of Public Health and Community Medicine, practical alternative in the study of diet and health University of Gothenburg, Sahlgrenska Academy, associations (Mai et al. 2005). Among other a priori Box 454, 405 30 Göteborg, Sweden e-mail: [email protected] patterns (such as the Healthy Eating Index or the Recommended Food Score), the Mediterranean diet E. Rothenberg A. Winkvist has already been shown to possess many beneficial Department of Clinical Nutrition, effects, as first became evident years ago in the Seven- University of Gothenburg, Sahlgrenska Academy, Box 459, 405 30 Göteborg, Sweden Country Study (Keys 1980). This diet is characterized 440 AGE (2011) 33:439–450 by a high intake of vegetables, legumes, fruits, nuts while the 70-year-old women born in those years and seeds, cereals (that in the past were largely were identified on the basis of previous inclusion in unrefined), and a high intake of olive oil combined the Prospective Population Study of Women in with a low intake of saturated lipids, a moderately high Gothenburg (PPSW, Bengtsson et al. 1997; Lissner intake of fish (depending on the proximity of the sea), et al. 2003). Response rate decreased from 84% and a low-to-moderate intake of dairy products (and then 86% for men and women, respectively, in the earliest mostly in the form of cheese and yogurt), a low intake birth cohort to 65% for both sexes in the most recent of meat and meat products as well as a regular but cohort. Overall, a total of 1,277 men and women still moderate intake of ethanol, primarily in the form of residing in the Gothenburg district had undergone diet wine and generally during meals (Willett et al. 1995). history examinations at the age of 70 (Eiben et al. The Mediterranean diet was first considered a 2004, 2005). Diet history was validated by comparison dietary pattern that protected against coronary heart of energy intake (EI) with estimated total energy disease (de Lorgeril et al. 1999). This was confirmed expenditure (TEE) by heart rate monitoring, activity very recently by a longitudinal study showing that diary, and doubly-labeled water (Rothenberg et al. Mediterranean diet is associated with a higher 1998). preservation of ventricular function and a more In order to exclude implausible dietary intakes, favorable prognosis after an acute coronary event subjects in the highest and lowest 5 percentiles of the (Chrysohoou et al. 2010). However, in previous ratio between energy intake and calculated basal studies, other beneficial effects on total mortality metabolic rate (BMR) were not included in the have been discovered (Trichopoulou et al. 2005). present analyses (N=149). BMR was predicted from Interestingly, a recent paper on the Swedish popula- weight (W) by means of standard equations for subjects aged 60–74 (males: 0.0499×W+2.93; females: tion showed a reduction in total mortality among young women (Lagiou et al. 2006). In addition, a 0.0386×W+2.875; Department of Health 1991). Other Mediterranean dietary pattern has also been shown to exclusions included cases missing information on diet increase longevity among European elderly of the or on potential confounders (N=91). The final study Healthy Aging: a Longitudinal study in Europe sample included 1,037 subjects (52.1% females). All (HALE) project (Knoops et al. 2004). cohorts have been monitored continuously for mortal- In the present study, we investigated the associa- ity by linking personal identification numbers with the tion of the Mediterranean diet pattern with total national death registration system. The most recent mortality in a population of Swedish 70 year olds. mortality follow-up was in 2009. In a sub-sample of 781 subjects, a follow up at 75 years for both weight and waist circumference was available. In supplemen- Subjects and methods tary analyses, observation time was truncated to 8.5 years to give all cohorts equal follow-up time Subjects, dietary assessment, and outcome definition when estimating survival probabilities. This study is based on the Gerontological and Definitions of food groups and of the Mediterranean Geriatric Population Studies in Gothenburg (H70; diet scores Eiben et al. 2004, 2005; Rothenberg et al. 1996, 1997, 1998). The Revenue Office of Gothenburg has a To assess the association of diet and lifestyle factors registry of all inhabitants, from which representative with total mortality, a refined version of the modified study groups of specific ages may be identified. Using Mediterranean diet score (refined mMDS) was calcu- this registry, population-based samples of 70-year-old lated, based on existing knowledge from the scientific residents were recruited for health examinations in literature on positive effects of wholegrain cereals 1971, 1981, 1992, and 2000. In the first two (Flight and Clifton 2006) and moderate alcohol intake examinations, 70-year-old cohorts born in 1901 and (Mukamal et al. 2010) as well as on the fact that 1911 were sampled and invited to participate based on polyunsaturated fats (PUFA) and not only monoun- date of birth. In the latter two examinations, men born saturated fats (MUFA) are the principal unsaturated in 1922 and 1930 were sampled in the same way, fats in non-Mediterranean diets (De Lorgeril et al. AGE (2011) 33:439–450 441 1994). The score comprised nine components: (1) the outcome of the iso-energetic replacement of 5% of vegetables and potatoes, (2) legumes, nuts, and seeds, energy provided by the reference macronutrient (the (3) fruit and fresh juices, (4) wholegrain cereals, (5) one not included in the model, e.g., carbohydrates) fish and fish products, (6) ratio of MUFA+PUFA to with a similar amount of calories provided by all the saturated fats (SAFA), (7) alcohol intake, (8) meat, other macronutrients included in the model (e.g., meat products, and eggs, (9) dairy products. Intake of proteins and fats). In these analyses, alcohol intake each component was adjusted to daily energy intakes was added separately on a g/day basis. of 2,500 kcal (10.5 MJ) for men and 2,000 kcal Next, analyses were performed both on the Medi- (8.5 MJ) for women. Although these values are not terranean diet scores (HALE and refined mMDS) on a considered as recommended daily energy intake for continuous scale as well as on scores of 6–9 in refined both genders, they were chosen based on previous mMDS or 5–8 in HALE mMDS in comparison with research on this subject (Knoops et al. 2004) and used the others. In order to examine if the estimates for to obtain energy adjusted associations in all the mortality were influenced by any of the covariates, or if analyses. The sex-specific median intakes were taken any of these could remove part of the effect, the crude as cut-off points. In any case, the final score is hazard ratios (HRs) for the dietary and lifestyle factors considered a measure of relative adherence to the were also calculated and compared with the adjusted Mediterranean diet pattern. The diet score varied from HRs. Finally, the mortality risk was also estimated in 0 (low-quality diet) to 9 (high quality diet). For each single food groups contributing to these scores. Food of the nine components, except for meat and dairy intakes were re-scaled according to medians, as done for products, a value of 1 was assigned to subjects whose the score calculations. For illustration, the dose– consumption was higher than the sex-specific median, response effect of the mortality risk in subjects with growing levels of refined mMDS compared to the risk and 0 to the others. For meat and dairy products, the reverse rule was applied. An alternative version of this in subjects characterized by a score ≤1 was investigated. score, previously used in the HALE project (HALE In addition, a series of sensitivity analyses were mMDS, Knoops et al. 2004) was also produced, performed, the most critical of which was the estima- including only MUFA in the numerator of the fat tion of the association between mortality risk and ratio, total instead of wholegrain cereals, and exclud- different versions of the refined mMDS, calculated by ing alcohol intake from the score. excluding, one at a time, each component of the score in order to determine that the effect was not driven by a Statistical analyses single food group. Other sensitivity analyses included adjusting for cardiovascular risk factors such as weight Cox proportional hazards models were used to and waist circumference change, baseline blood pres- estimate the effects of diet, smoking status, education, sure, fasting glucose, total cholesterol, and triglycerides physical activity, and other potential confounders on or for activities of daily living variable (ADL, based on total mortality, to November 2009. All the models used need of help for everyday activities). The activities of in this study were adjusted for potential confounders daily living variable was based on the Katz index, such as sex, baseline body mass index (BMI), waist which is similar to the Barthel Index, and hence based circumference, smoking status, physical activity level, on the need of help for at least one out of nine different marital status, education, and birth cohort. activities, including house cleaning, shopping, transport, Analyses of macronutrients and mortality were first cooking, washing, dressing, using the toilet, ambulation, performed by means of the standard multivariate model and feeding (Sonn and Asberg 1991). Furthermore, (Willett 1998), in which each macronutrient was tested analyses were repeated after excluding the first 2 years separately, with or without adjustment for energy of total mortality follow-up. intake. Further analyses were conducted by means of We also repeated the Cox analysis on the refined the multivariate nutrient density model (Willett 1998), mMDS, by censoring at a fixed follow-up duration of in which the nutrient densities (percentages of energy 8.5 years (approximately the longest follow-up time from non-alcohol sources in a scale of 5% units) from in the latest-born cohort) instead of censoring at a total fat and protein intakes are included as indepen- fixed date. Crude proportions of subjects still alive at dent variables. The model thus estimates the effect on 8.5 years were also calculated according to three 442 AGE (2011) 33:439–450 different levels of the score (≤3, 4–6, ≥7) and the consistently high (1.5±0.3), indicating a good quality association between adherence to the Mediterranean of nutritional data. In the subjects excluded for diet pattern and survival was assessed by means of a implausible dietary intakes (highest and lowest 5 chi-square test. percentiles of energy intake vs. basal metabolic rate Finally, potential confounders (see above) were ratio), EI/BMR was either less than 0.9 or greater than examined in more detail, including an estimation of 2.2 (data not shown). the mortality HRs in subjects characterized by a score Regarding the two scores measuring the adherence to in the highest four levels (>4 for the HALE mMDS, the Mediterranean diet (Table 2), no major differences >5 for the refined mMDS) compared to the other appeared from cohort to cohort, although for the subjects. Moreover, the assessment of the effect of all refined mMDS a slight but significant increase was main covariates on mortality included in the previous evident (4.3±1.6 in 1901 cohort to 4.8±1.8 in 1930 models was also performed. cohort), probably due to the increase in wholegrain, All statistical analyses were considered significant alcohol, and PUFA intakes included in this score and under a p value of 0.05 and were carried out using the not in the HALE mMDS. SAS statistical software version 9.2 (SAS Institute, Table 3 shows the descriptive analyses of potential Cary, NC, USA). confounders. Both genders were equally represented in the population (47.9% men). Subjects characterized Bioethics by a BMI <20 were only a minority (3.1%), while obese subjects constituted 15.8% of the total popula- The latest H70 examinations (since 1992) have tion. A high waist circumference (>88 cm in women, been approved by the Gothenburg University Ethics >102 cm in men), characterized 31.0% of the subjects. Only 14.2% of subjects were physically Committee in accordance with the Declaration of Helsinki (1989) of the World Medical Association. inactive, 62.1% were married, and more than 30% of All participants were informed of the aims and subjects were smokers (or stopped less than 10 years procedures of the study and gave their consent. ago) or had an education above basic (i.e., 6 years of schooling or more). Overall, in the studied popula- tion, there were 630 deaths (60.8%) until November Results 2009: 309 women and 321 men. Regarding main cardiovascular risk factors depicted Descriptive analyses in Table 3, which include variations in anthropometry over 5 years, we observed a decrease in mean weight Main descriptive analyses are reported in Tables 1 from 74.2 kg±12.9 at baseline to 72.3 kg±13.2 at (dietary variables), 2 (Mediterranean diet scores), and 3 follow up (mean change: −1.9 kg±5.4), a quite stable (potential confounders and other cohort descriptors). waist circumference (mean change: −0.2 cm±7.4) and Dietary intake distributions across the birth cohorts BMI (mean change: −0.3 kg/m ±1.9). Diastolic blood have previously been described (Eiben et al. 2004; pressure means were less than 90 mmHg (87.4 mmHg± Rothenberg et al. 1996). Table 1 shows that, according 11.9), while systolic blood pressure mean was above to trend test, across the four birth cohorts there was an 140 mmHg (159.0 mmHg±23.6). Regarding plasma increase in the consumption of vegetables and potatoes, parameters (only available at baseline), fasting fruit, legumes plus nuts and seeds, fish and fish glucose was equal to 100.8 mg/dL±36.0, total choles- products, meat and meat products, red wine and terol was 233.3 mg/dL±77.8 while triglycerides was alcohol. In contrast, a decrease in the consumption of 132.4 mg/dL±70.6. cereals (both total and wholegrain) was observed. No clear patterns emerged for dairy products, MUFA/ Diet quality and mortality SAFA ratio and (MUFA+PUFA)/SAFA ratio. Regarding macronutrient intakes, a trend in protein, carbohydrate, In the present cohort, the total mortality was around and fat intake was observed. Energy intakes showed a 60%, with a higher rate in the earlier born cohorts and slight tendency to an increase across the cohorts, and a lower in the later born cohorts (100% in 1901, 98% the ratio between energy intake and BMR was in 1911, 48% in 1922, and 15% in 1930 birth cohort). AGE (2011) 33:439–450 443 Table 1 Means and standard deviation in comparison with medians and 95% Confidence Limits (CLs) of dietary variables used in the analyses, including test for trend across the birth cohorts e,f Food groups/macronutrients Overall mean (N=1,037) Median intakes and 95% CLs p for trend across birth cohorts a,b Vegetables and potatoes (g/day) 237.6±98.7 ♀: 209.5 (99.6; 406.3) <0.0001 ♂: 239.0 (120.1; 432.2) a,b Fruit (g/day) 196.6±146.3 ♀: 176.4 (22.3; 527.7) <0.0001 ♂: 155.5 (14.0; 456.0) a,b Legumes nuts and seeds (g/day) 15.2±20.2 ♀: 2.0 (0; 40.0) <0.0001 ♂: 13.3 (0; 60.0) Cereals (g/day) 207.4±104.0 ♀: 165.0 (68.2; 383.0) (−)<0.0001 ♂: 213.0 (98.3; 442.1) Wholegrain cereals (g/day) 107.9±95.6 ♀: 74.2 (0; 298.5) (−)<0.0001 ♂: 92.8 (0; 322.0) a,b Fish and fish products (g/day) 53.8±35.8 ♀: 45.2 (12.8; 105.3) <0.0001 ♂: 53.7 (13.8; 129.5) a,b Dairy products (g/day) 445.1±251.7 ♀: 373.3 (127.9; 829.5) 0.41 ♂: 446.0 (74.0; 1061.9) Meat and meat products (g/day) 105.4±47.9 ♀: 89.7 (38.4; 168.8) <0.0001 ♂: 109.1 (52.5; 204.9) Meat, meat products, eggs (g/day) 129.5±55.3 ♀: 110.1 (47.9; 187.3) <0.001 ♂: 137.7 (66.3; 251.1) Red wine (g/day) 2.0±4.7 ♀: 0 (0; 9.8) <0.001 ♂: 0 (0; 10.7) Alcohol (g/day) 6.0±9.1 ♀: 1.3 (0; 13.9) <0.0001 ♂: 5.3 (0; 28.5) Carbohydrate (g/day) 249.1±64.8 ♀: 217.1 (143.5; 316.8) 0.01 ♂: 272.4 (192.8; 398.2) Protein (g/day) 80.1±20.0 ♀: 71.2 (48.6; 100.4) <0.0001 ♂: 85.2 (58.3; 129.0) Fat (g/day) 86.1±25.4 ♀: 74.6 (46.6; 110.) <0.0001 ♂: 93.5 (60.1; 146.5) MUFA/SAFA 0.8±0.2 ♀: 0.8 (0.6; 1.1) 0.40 ♂: 0.8 (0.6; 1.1) (MUFA+PUFA)/SAFA 1.2±0.3 ♀: 1.1 (0.8; 1.6) 0.50 ♂: 1.1 (0.7; 1.7) Energy intake (100 kcal) 21.4±5.0 ♀: 18.5 (13.3; 25.4) 0.04 ♂: 23.6 (17.7; 33.6) Energy/BMR 1.5±0.3 ♀: 1.4 (1.0; 1.9) 0.30 ♂: 1.4 (1.1; 2.1) Included in HALE mMDS Included in refined mMDS Included in place of alcohol in an additional analysis Not included in any scores, just included for descriptive purposes Birth–cohort effect from a regression model adjusted for gender, BMI, waist circumference, physical activity, smoking status, marital status and education All significant trends positive in direction except if indicated with (−) In the first part of the present work, we analyzed the intake (adjusted or unadjusted for energy intake) in association between diet macronutrient composition standard multivariate models. Alcohol was also not and total mortality. No clear association emerged associated on a continuous scale with total survival either when analyzing protein, fat, and carbohydrate time (Table 4). Similar results were obtained when 444 AGE (2011) 33:439–450 Table 2 General descriptive of Mediterranean diet score distributions, both stratified by birth cohort and for the overall population Mediterranean diet score 1901 (N=323) 1911 (N=214) 1922 (N=88) 1930 (N=412) Overall (N=1,037) p for trend across birth cohorts Refined mMDS (mean±SD) 4.3±1.6 4.0±1.6 4.8±1.5 4.8±1.8 4.5±1.7 <0.001 Medians (5th;95th perc) 4 (2; 7) 4 (1; 7) 5 (2; 7) 5 (2; 8) 4 (2; 7) HALE mMDS (mean±SD) 4.1±1.5 3.6±1.4 4.1±1.4 4.1±1.6 4.0±1.5 0.02 Medians (5th;95th perc) 4 (2; 6) 4 (1; 6) 4 (2; 6) 4 (1; 7) 4 (2; 6) Birth–cohort effect from a regression model adjusted for gender, BMI, waist circumference, physical activity, smoking status, marital status, and education employing the nutrient density model to describe the In light of the null results obtained by the analyses effect of the reciprocal substitution of 5% of energy of macronutrients, we then focused on the identifica- intake from each macronutrient with carbohydrates tion of a possible dietary pattern, which could be (Table 4). related to total mortality. In particular, we chose to study the Mediterranean diet pattern by means of a Table 3 Main covariate frequencies and cohort description refined version of the modified Mediterranean Diet Potential confounders Frequency (%) Score (refined mMDS), which we consider to reflect a closer adherence to the classic Mediterranean diet Total sample (N = 1,037) pattern. Male gender 47.9 Table 5 shows the association of refined mMDS, in BMI<20 3.1 comparison to the previously used HALE mMDS BMI>30 15.8 (Knoops et al. 2004) with mortality risk, as well as the High waist circumference 31.0 same outcome for those food groups on whose intakes Low physical activity at 70 14.2 the adherence to Mediterranean diet pattern was Married at age 70 62.1 assessed. Smoker/stopped <10 years ago 30.6 Regarding Mediterranean diet score, an inverse School education above basic 30.5 association with total mortality was shown for the Activities of daily living (ADL) 12.7 continuous refined mMDS (HR=0.93, 95% CI: 0.89; 0.98), while no significant association emerged with Additional covariates Means±SD HALE mMDS, although the trend was toward an Baseline diastolic blood pressure (mmHg) 87.4±11.9 inverse association (HR=0.97, 95% CI: 0.92; 1.02). Baseline systolic blood pressure (mmHg) 159.0±23.6 Moreover, the comparison of the lowest-risk group Baseline fasting glucose (mg/dL) 100.8±36.0 Baseline total cholesterol (mg/dL) 233.3±77.8 Table 4 Hazard ratios and 95% confidence limits from a Cox- Baseline triglycerides (mg/dL) 132.4±70.6 proportional hazard model (adjusted for gender, baseline BMI, waist circumference, physical activity, marital status, smoking status, birth cohort, and education) estimating the association of Subsample with follow up at 75 years (N = 781) macronutrient with mortality risk and different substitution of Changes in anthropometry Means±SD energy from each macronutrient with the same amount of energy of another one Weight change (kg) −1.9±5.4 Baseline 74.2±12.9 Macronutrient (10 g) HR 95% CLs Follow up 72.3±13.2 Waist circ. change (cm) −0.2±7.4 Alcohol 0.98 0.86; 1.11 Baseline 90.0±11.4 Protein 0.97 0.90; 1.04 Follow up 90.0±11.8 CHO 1.00 0.98; 1.03 BMI change (kg/height ) −0.3±1.9 Fat 1.00 0.94; 1.07 Baseline 26.4±3.9 From CHO to protein 5% energy 0.95 0.80; 1.14 Follow up 26.1±4.1 From CHO to fat 5% energy 1.01 0.94; 1.09 AGE (2011) 33:439–450 445 Table 5 Hazard ratios and 95% confidence limits from a Cox- 0.71; 0.98). In contrast, other food groups were not proportional hazard model (adjusted for gender, baseline BMI, significantly associated with mortality, such as vege- waist circumference, physical activity, marital status, smoking tables and potatoes (HR=1.06, 95% CI: 0.90; 1.24), status, birth cohort, and education) estimating the association of both HALE and refined mMDS with mortality risk fruit (HR=1.03, 95% CI: 0.87; 1.21), legumes plus nuts and seeds (HR=0.98, 95% CI: 0.83; 1.16), and Mediterranean diet score or food group HR 95% CLs fish (HR=0.96, 95% CI: 0.82; 1.13). No association was shown for either of the fat ratios (MUFA/SAFA Refined mMDS 0.93 0.89; 0.98 ratio: HR=0.98, 95% CI: 0.84: 1.15; (MUFA+PUFA)/ Crude estimate 0.92 0.88; 0.97 SAFA ratio: HR=0.96, 95% CI: 0.82; 1.13). It is worth Highest 4 levels vs. the others 0.82 0.67; 0.99 mentioning that vegetables alone and potatoes alone Crude estimate 0.81 0.67; 0.99 did not show any association with total mortality (data HALE mMDS 0.97 0.92; 1.02 not shown). Crude estimate 0.97 0.92; 1.03 In Fig. 1 dose–response analysis results are Highest 4 levels vs. the others 0.94 0.79; 1.11 depicted. Briefly, the group characterized by a refined Crude estimate 0.94 0.80; 1.11 mMDS equal to 0 or 1 was the reference, compared to High intake/level of: the other groups with increasing levels of the score. Vegetables and potatoes 1.06 0.90; 1.24 Although none of the single comparisons reached Fruit 1.03 0.87; 1.21 statistical significance, the analysis showed a dose– Legumes, nuts, and seeds 0.98 0.83; 1.16 response tendency, thus suggesting a decrease in the Cereals 1.01 0.86; 1.19 mortality risk as long as the adherence to the Wholegrain cereals 0.85 0.73; 1.00 Mediterranean diet pattern increased. Fish 0.96 0.82; 1.13 Alcohol 0.77 0.61; 0.97 Supplementary analyses MUFA/SAFA ratio 0.98 0.84; 1.15 (MUFA+PUFA)/SAFA ratio 0.96 0.82; 1.13 A sensitivity analysis was performed on the refined Low intake of: mMDS, with the aim of understanding whether any Dairy products 0.82 0.70; 0.96 one of the food group items included in the score Meat and meat products 0.89 0.76; 1.05 could invalidate the effect of the entire score, with Meat, meat products and eggs 0.84 0.71; 0.98 respect to mortality. Nine different scores were then produced, each excluding one item at a time (and keeping all others). Table 6 clearly demonstrates that (highest four levels of the score) versus the other the association of the refined mMDS with mortality is robust and survives all the item-by-item exclusions. subjects showed a significant inverse association for the refined mMDS (HR=0.82, 95% CI: 0.67; 0.99), but Furthermore, in order to test whether the positive not for the highest levels of the HALE mMDS (HR= effect of alcohol intake within the score could be 0.94, 95% CI: 0.79; 1.11; Table 5). Crude estimates obtained considering red wine (instead of total alcohol were also calculated and, as shown in the table, they are intake from all alcoholic beverages), considered a more quite overlapping with the adjusted ones, thus showing accurate indicator of the Mediterranean alcoholic that covariates did not have any strong influence on drinks, we ran an additional analysis, ending up with results. The protective effect of the Mediterranean diet a similar outcome and a slightly stronger hazard ratio pattern was stronger in the two youngest birth cohorts (HR: 0.92, 95% CI: 0.87; 0.97; Table 6). compared to the others (data not shown). In order to confirm that the effect of diet on mortality When studying the effect of single food groups, an was independent of the subjects’ functional capacity levels, and thus that the effect of Mediterranean diet inverse association with total mortality was shown for high intakes of wholegrain cereals (HR=0.85, 95% was not entirely explained by how impaired the CI: 0.73; 1.00), alcohol consumption (HR=0.77, 95% subjects were at the time of diet assessment we also CI: 0.61; 0.97) and low intake of dairy products (HR= repeated all the models adjusting for activities of daily 0.82, 95% CI: 0.70; 0.96) as well as low intakes of living variable and found no substantial change in meat, meat products, and eggs (HR=0.84, 95% CI: results. Similar analyses were performed adjusting for 446 AGE (2011) 33:439–450 Fig. 1 Dose–response analyses based on Cox- proportional hazard model (adjusted for gender, baseline BMI, waist circumference, physical activity, marital status, smoking status, birth cohort and education) comparing mortality risk in subjects characterized by a refined mMDS ≤1 with the same risk in subjects in the other score categories. Dots and vertical lines indicate HR and 95% confidence limits other cardiovascular risk factors, i.e., baseline blood more, to minimize the possibility that diet or lifestyle pressure, fasting glucose, total cholesterol, and trigly- factors had changed in response to subclinical diseases, cerides; this did not strongly affect the results, except all analyses were also repeated after exclusion of the that the association of the dichotomous score (higher vs. subjects who had died in the first 2 years of follow up. lower) was attenuated (HR=0.85, 95% CI: 0.70; 1.04). Here again, no clear difference emerged, thus confirm- Adjustment for weight or waist circumference change ing our main results. did not substantially affect the association of the refined Finally, the association of the refined mMDS with mMDS with mortality. However, both were inversely total mortality was also tested in a Cox model in associated with total mortality (weight change: HR= which subject censoring was done at a fixed follow up 0.95, 95% CI: 0.93; 0.97; waist circumference change: time (8.5 years, the maximum in the youngest birth HR=0.98, 95% CI: 0.97; 1.00), suggesting a greater cohort) instead at a fixed date. The previous results mortality risk in association with weight loss. Further- were confirmed although somewhat attenuated (HR= 0.92, 95% CI: 0.86; 1.00). The distribution of crude proportions of survival rates at 8.5 years, stratified by increasing levels of the refined mMDS were as Table 6 Hazard ratios and 95% confidence limits from a Cox- proportional hazard model-based sensitivity analysis estimating follows: 67% survival among those characterized by the effect of the exclusion of each level of the refined mMDS+ a score ≤3,77% survival at intermediate levels of the alcohol on mortality risk (adjusted for gender, baseline BMI, score (from 4 to 6) and 80% survival among the waist circumference, physical activity, marital status, smoking subjects with a score ≥7(p=0.02). status, birth cohort, and education) Mediterranean diet score HR 95% CLs Effect modification and covariates associated with diet and mortality Refined mMDS 0.93 0.89; 0.98 Without fat ratio 0.92 0.87; 0.97 In order to identify potential effect modifiers, statis- Without vegetables and potatoes 0.91 0.87; 0.96 tical interactions between all dietary and potential Without fruit 0.92 0.87; 0.97 confounders were tested, but none of them were Without legumes, nuts and seeds 0.93 0.88; 0.97 statistically significant at a 0.05 level. The association Without cereals 0.94 0.89; 0.99 of the highest levels of the refined mMDS with each Without fish 0.92 0.87; 0.97 of the above-mentioned confounders was also inves- Without dairy 0.94 0.89; 0.99 tigated (Table 7). Subjects ranked to the highest levels Without meat 0.94 0.90; 0.99 of the score were less likely to be inactive (OR=0.42; Without alcohol 0.94 0.89; 0.99 95% CI: 0.25; 0.71) and more likely to be married Refined mMDS (including red wine 0.92 0.87; 0.97 (OR=1.47, 95% CI: 1.07; 2.01), more educated instead of alcohol) (OR=1.38, 95% CI: 1.02; 1.86) or born within the AGE (2011) 33:439–450 447 Table 7 Hazard ratios and Confounders Association with highest Association with 95% confidence limits from levels of refined mMDS mortality a Cox-proportional hazard model estimating the OR 95% CLs HR 95% CLs association of potential confounders included in the Male gender 0.90 0.66; 1.23 2.02 1.67; 2.44 final model and mortality risk in the whole studied BMI<20 1.06 0.45; 2.50 1.32 0.90; 1.96 population (N=1,037) BMI>30 0.92 0.56; 1.51 0.94 0.72; 1.24 High waist circumference 0.80 0.54; 1.18 1.29 1.04; 1.60 Low physical activity at 70 0.42 0.25; 0.71 1.29 1.04; 1.59 Married at age 70 1.47 1.07; 2.01 0.78 0.66; 0.93 Smoker/stopped <10 years ago 1.07 0.77; 1.47 1.37 1.14; 1.64 School education above basic 1.38 1.02; 1.86 0.91 0.75; 1.11 Birth cohort 1.03 1.01; 1.04 0.97 0.97; 0.98 more recent birth cohorts (OR 0 1.03, 95% CI: the time of the dietary assessment, or biased in some 1.01; 1.04). way by how impaired they were at enrolment. Table 7 also shows the association of the potential Furthermore, the effect was only slightly attenuated confounders included in the final model with total when adjusting for several indicators of cardiovascu- mortality. As expected, there is a positive association lar risk, suggesting that the protective effect was not between the risk of mortality and male gender (HR= limited to high-risk subjects. 2.02, 95% CI: 1.67; 2.44), high waist circumference Other studies had already found a positive associ- (HR=1.29, 95% CI: 1.04; 1.60), low level of physical ation of Mediterranean diet and health (Lagiou et al. activity (HR=1.29, 95% CI: 1.04; 1.59) and smoking 2006; Knoops et al. 2004; Trichopoulou et al. 1995), (HR=1.37, 95% CI: 1.14; 1.64). Inverse associations showing the effects of increasing the intake of were found for marital status (HR=0.78, 95% CI: vegetables and fruit, fish, and cereals, while decreas- 0.66; 0.93), and for birth cohort, which we considered ing animal products such as dairy and meat. In our as a proxy for age/period effects (HR=0.98, 95% CI: study, we started by reproducing the same mMDS that 0.97; 0.98). Finally, BMI did not show any significant was previously created by the HALE working group association. for their paper on Mediterranean diet and total mortality (Knoops et al. 2004) and did not find a significant trend, even when testing the association at the highest Discussion levels. However, we found an inverse association with total mortality when a refined version of the mMDS In this paper, we investigated the association of a was tested instead of the original score. The associa- dietary pattern close to the Mediterranean diet with tion was indeed strong and robust after calculating the total mortality in population-based cohorts of 70-year- score considering wholegrain cereals instead of total old Swedes. Although we did not find an association cereals, adding alcohol (or red wine) intake, consider- when testing the previously defined HALE mMDS on ing egg intake together with meat products and our Swedish population (Knoops et al. 2004), we calculating the fat ratio including PUFA and not only were able to refine this score in order to obtain one MUFA in the numerator. The addition of alcohol that best described strict adherence to a Mediterranean intake should not be considered problematic, since the intakes in this population were not high (75th diet-like pattern. None of the items included in the score were found to be essential for the association percentile: 7.6 g/day, 90th percentile: 16.3 g/day) and with the total score. Finally, the association was not can then be considered in an order of magnitude quite attenuated when adjusting for ADL, thus showing that close to international recommendations. Interestingly, the protection by the Mediterranean pattern was not beneficial effects of moderate alcohol intake have limited to those subjects still living independently at been recently confirmed in a study with a follow up of 448 AGE (2011) 33:439–450 20 years that also considered many confounders estimated intakes with National Surveys; Becker typically associated with abstaining (such as past 1994), the possibility of adjusting for different known history of heavy drinking behavior, Holahan et al. confounders and the availability of weight and waist 2010). Furthermore, our association was also con- circumference measurements both at baseline and at firmed when including red wine instead of alcohol, follow up. The limitations include the lack of repeated confirming that the positive result characterizes the dietary assessment and the small size of the study subjects adhering to the healthiest diet patterns. population. Moreover, it must be acknowledged that The possible explanation of the necessity of improv- diet patterns alone may inherently co-vary with other ing an existing score that had already worked in health-related phenomena (e.g., healthy lifestyles, previous studies to find a robust association in the illnesses, weight status). Thus, it is important to note Swedish elderly may be due to the obvious differences that associations were independent of cardiovascular between the original Mediterranean diet and the risk factors, weight and waist circumference change, Swedish Mediterranean-like diet. This is particularly SES, education, ADL, and other risk factors. More- important in our population, since many subjects were over, it is worth mentioning that although the analyses born at the beginning of last century, when many were always adjusted for birth cohort, the results products that were common in Southern Europe were could still be influenced by a residual cohort effect as not consumed in large quantities in the north. Swedish suggested by the fact that the protective effect of the food habits have indeed undergone major changes in the Mediterranean diet pattern was stronger in the last 30 years, particularly in relation to consumptions of youngest cohorts. This was probably due to a higher fresh fruit and vegetables (now available in quantity quality of nutritional data (or higher health conscious- even on the Scandinavian market) and cereals as well as ness in the subjects) in the most recent surveys. From the above mentioned results, it clearly a higher unsaturated fat proportion (Eiben et al. 2004). Another reasonable example is represented by MUFA emerges that it would be useful to put more emphasis sources, mainly olive oil in Southern European on dietary recommendations directed to the elderly in countries, mostly margarines (a source of trans fatty order to encourage increased consumption of fruit and acids, especially at the time they were measured) in the vegetables, wholegrain cereals and fish, while reduc- oldest birth cohorts and only marginally derived from ing the intake of dairy and meat products, in favor of olive oil in the 1930 birth cohort. other healthier protein sources such as legumes. This However, by applying stricter rules, such as is particularly important in Sweden, considering that substituting wholegrain cereals instead of total cereals current guidelines do not always place sufficient (often characterized by a higher sugar content than in emphasis on this type of recommendation. For South Europe), including PUFA intake (in Sweden, instance, although our results are quite in line with fish is an important contributor), adding alcohol (and Nordic Nutrition Recommendations (Nordic Council also red wine only), and including egg intake (a of Ministers 2004), the present Swedish Nutrition probable indicator of a low-quality diet), we found a Recommendations Objectified (Barbieri and Lindvall protective pattern in our Swedish population. 2005), is in some ways different from Mediterranean Another important issue is that single food groups diet-based features. Indeed, the former considers were not always found to be inversely or directly acceptable a daily intake of 500 ml milk (in addition associated with the mortality risk, although the full to other dairy products) and suggests some equilibrated score is robust to exclusions of any single component. menus that do not consider olive oil either as a This suggests that the use of this and similar scores in condiment or as a cooking fat, while including nutritional epidemiology studies is indeed a useful margarine on a daily basis. Moreover, it often refers to strategy for investigating associations between diet “refined rye bread” and “white bread” instead of and health outcomes. suggesting wholegrain products, and includes meat Our study has both strengths and limitations, the products in two meals per day. At the same time, it is former being the high quality of nutritional data also worth mentioning that carbohydrate restriction is obtained by a validated diet history during a face to becoming an increasingly popular (although not scien- face interview with the dietician (confirmed by the tifically based) weight control method in the popula- high EI/BMR ratio and by concordance of our tion, a modification which is also not supported, vis-à- AGE (2011) 33:439–450 449 vis mortality, in our nutrient-level analyses simulating References carbohydrate replacement. Overall, studies of dietary patterns are inherently Appel LJ (2008) Dietary patterns and longevity. Circulation complex. However, regardless of scientific approach, 118:214–215 there is a remarkable convergence of evidence on the Barbieri HE, Lindvall C (2005) De svenska näringsrekommen- dationerna översatta till livsmedel: underlag till generella fact that dietary patterns associated with longevity råd på livsmedels-och måltidsnivå för friska vuxna emphasize fruits and vegetables and are reduced in Swedish Nutrition Recommendations Objectified (SNO). saturated fat, meats, refined grains, sweets, and full- Basis for general advice on food consumption for healthy fat dairy products (Appel 2008). This is not only the adults. Livsmedelsverkets rapport n. 20/2005. http://www. slv.se/upload/dokument/rapporter/mat_naring/ case for the Mediterranean diet pattern, but also for Report_20_2005_SNO_eng.pdf other pattern such as the DASH diet (Parikh et al. Becker BW (1994) Dietary habits and nutrient intake in 2009) or the Okinawa diet (Willcox et al. 2009). Sweden 1989. Statens livsmedelsverk, Uppsala, 240 pp Bengtsson C, Ahlqvist M, Andersson K, Björkelund C, Lissner L, Söderström M (1997) The prospective population study of women in Gothenburg, Sweden, 1968–69 to 1992–93. Conclusions A 24-year follow-up study with special reference to participation, representativeness, and mortality. Scand J To conclude, we can reasonably state that the Prim Health Care 15:214–219 Chrysohoou C, Panagiotakos DB, Aggelopoulos P, Kastorini adherence to a Mediterranean-like dietary pattern is CM, Kehagia I, Pitsavos C, Stefanadis C (2010) The inversely related to total mortality also in a Swedish Mediterranean diet contributes to the preservation of left population of elderly subjects. Our hope is that the ventricular systolic function and to the long-term favorable results of the present research will stimulate a prognosis of patients who have had an acute coronary event. Am J Clin Nutr 92:47–54 productive discussion on these issues and be consid- De Lorgeril M, Renaud S, Mamelle N, Salen P, Martin JL, ered in updated food- and nutrient-level guidelines for Monjaud I, Guidollet J, Touboul P, Delaye J (1994) the population. Mediterranean alpha-linolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet 343:1454–1459 de Lorgeril M, Salen P, Martin JL, Monjaud I, Delaye J, Acknowledgment The authors’ responsibilities were as fol- Mamelle N (1999) Mediterranean diet, traditional risk lows—GT performed the data analysis and wrote the manu- factors, and the rate of cardiovascular complications after script; LL coordinated the research, contributed to the myocardial infarction: final report of the Lyon Diet Heart interpretation of results and to the writing of the manuscript; Study. Circulation 99:779–785 ER and GE provided support for the correct application of Department of Health (1991) Report on health and social nutritional data, helped with the interpretation of results and subjects; 41. Dietary reference values for food energy and gave critical comments on the manuscript; AW contributed to nutrients for the United Kingdom. HMSO, London the interpretation of results and gave critical comments on the Eiben G, Andersson CS, Rothenberg E, Sundh V, Steen B, manuscript; VS provided statistical expertise and gave critical Lissner L (2004) Secular trends in diet among elderly comments on the manuscript. The research was funded by the Swedes—cohort comparisons over three decades. Public Swedish Council on Working Life and Social Research (FAS) Health Nutr 7:637–644 EpiLife centre. Eiben G, Dey DK, Rothenberg E, Steen B, Björkelund C, Bengtsson C, Lissner L (2005) Obesity in 70-year-old Financial disclosures All authors have no financial disclosures. Swedes: secular changes over 30 years. Int J Obes (Lond) 29:810–817 Flight I, Clifton P (2006) Cereal grains and legumes in the Conflict of interest None declared. prevention of coronary heart disease and stroke: a review of the literature. Eur J Clin Nutr 60:1145–1159 Haveman-Nies A, de Groot L, van Staveren W (2003) Relation Disclaimer The views expressed are those of the authors and of dietary quality, physical activity, and smoking habits to should not be construed to represent the positions of anybody 10-year changes in health status in older Europeans in the else. SENECA study. Am J Public Health 93:318–323 Huijbregts P, Feskens E, Rasanen L, Fidanza F, Nissinen A, Menotti A, Kromhout D (1997) Dietary pattern and 20 year Open Access This article is distributed under the terms of the mortality in elderly men in Finland, Italy, and the Nether- Creative Commons Attribution Noncommercial License which lands: longitudinal cohort study. BMJ 315:13–17 permits any noncommercial use, distribution, and reproduction Keys AB (1980) Seven countries: a multivariate analysis of in any medium, provided the original author(s) and source are death and coronary heart diseases. Harvard University credited. Press, Cambridge 450 AGE (2011) 33:439–450 Knoops KT, de Groot LC, Kromhout D, Perrin AE, Moreiras- Rothenberg E, Bosaeus I, Steen B (1997) Evaluation of energy Varela O, Menotti A, van Staveren WA (2004) Mediterra- intake estimated by a diet history in three free-living nean diet, lifestyle factors, and 10-year mortality in elderly 70 year old populations in Gothenburg, Sweden. Eur J European men and women: the HALE project. JAMA Clin Nutr 51:60–66 292:1433–1439 Rothenberg E, Bosaeus I, Lernfelt B, Landahl S, Steen B Holahan CJ, Schutte KK, Brennan PL, Holahan CK, Moos BS, (1998) Energy intake and expenditure: validation of a diet Moos RH (2010) Late-life alcohol consumption and 20-year history by heart rate monitoring, activity diary and doubly mortality. Alcohol Clin Exp Res 34:1–11 labeled water. Eur J Clin Nutr 52:832–838 Lagiou P, Trichopoulou D, Sandin S, Lagiou A, Mucci L, Wolk Seymour J, Calle E, Flagg E, Coates R, Ford E, Thun M (2003) A, Weiderpass E, Adami HO (2006) Mediterranean dietary Diet quality index as a predictor of short-term mortality in pattern and mortality among young women: a cohort study the American Cancer Society Cancer Prevention Study II in Sweden. Brit J Nutr 96:384–392 Nutrition Cohort. Am J Epidem 157:980–988 Lissner L, Skoog I, Andersson K, Beckman N, Sundh V, Waern Sonn U, Asberg KH (1991) Assessment of activities of daily M, Zylberstein DE, Bengtsson C, Björkelund C (2003) living in the elderly. A study of a population of 76-year- Participation bias in longitudional studies—experiences olds in Gothenburg, Sweden. Scand J Rehabil Med from the The Population Study of women in Gothenburg. 23:193–202 Scand J Prim Health Care 21:242–247 Trichopoulou A, Kouris-Blazos A, Wahlqvist ML, Gnardellis Mai V, Kant AK, Flood A, Lacey JV Jr, Schairer C, Schatzkin C, Lagiou P, Polychronopoulos E, Vassilakou T, Lipworth A (2005) Diet quality and subsequent cancer incidence L, Trichopoulos D (1995) Diet and overall survival in and mortality in a prospective cohort of women. Int J Epid elderly people. BMJ 311:1457–1460 34:54–60 Trichopoulou A, Orfanos P, Norat T, Bueno-de-Mesquita B, Mukamal KJ, Chen CM, Rao SR, Breslow RA (2010) Alcohol Ocke MC, Peeters PH et al (2005) Modified Mediterra- consumption and cardiovascular mortality among US nean diet and survival: the EPIC-Elderly study. BMJ adults, 1987 to 2002. J Am Coll Cardiol 55:1328–1335 330:991 Nordic Council of Ministers (2004) Nordic Nutrition Recom- Willcox DC, Willcox BJ, Todoriki H, Suzuki M (2009) The mendations 2004. Integrating nutrition and physical Okinawan diet: health implications of a low-calorie, activity. Norden, Copenhagen nutrient-dense, antioxidant-rich dietary pattern low in Parikh A, Lipsitz SR, Natarajan S (2009) Association between glycemic load. J Am Coll Nutrition 28:500S–516S a DASH-like diet and mortality in adults with hyperten- Willett WC (1998) Nutritional epidemiology, 2nd edn. Oxford sion: findings from a population-based follow-up study. University Press, New York Am J Hypertens 22:409–416 Willett WC, Sacks F, Trichopoulou A, Drescher G, Ferro-Luzzi Rothenberg E, Bosaeus I, Steen B (1996) Food habits in three A, Helsing E, Trichopoulos D (1995) Mediterranean diet 70-year-old free-living populations in Gothenburg, Sweden. pyramid: a cultural model for healthy eating. Am J Clin A 22-year cohort study. Scand J Nutr 40:104–110 Nutr 61:1402S–1406S

Journal

AgePubmed Central

Published: Nov 26, 2010

There are no references for this article.