Eur J Nutr DOI 10.1007/s00394-014-0811-z
ORIGINAL CONTRIBUTION
Mediterranean diet and cognitive decline over time in an elderly Mediterranean population Antonia Trichopoulou · Andreas Kyrozis · Marta Rossi · Michalis Katsoulis · Dimitrios Trichopoulos · Carlo La Vecchia · Pagona Lagiou
Received: 3 June 2014 / Accepted: 2 December 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Evidence suggests that dietary patterns compatible with the traditional Mediterranean diet (MD) may protect against cognitive decline. We prospectively assessed whether adherence to MD in the Mediterranean country of Greece is inversely associated with cognitive decline in the elderly and whether any particular MD component may play a key role. Methods Elderly men and women (N = 401) residing in the greater Athens area had dietary variables ascertained in 1994–1999. Adherence to MD was represented by the MD score [MDS, 0–3 (low), 4–5 (intermediate), 6–9 (high)]. The mini-mental state examination (MMSE) was administered by trained professionals to individuals aged 65 years or older in 2004–2006 (first assessment) and Electronic supplementary material The online version of this article (doi:10.1007/s00394-014-0811-z) contains supplementary material, which is available to authorized users.
re-administered in 2011–2012 (second assessment). MMSE change (cMMSE) was categorized as: improved/unchanged (cMMSE ≥ 0), mildly lower (cMMSE −1 to −4) or substantially lower (cMMSE ≤ −5). Associations were evaluated through multinomial logistic regression. Results Decline in MMSE performance was inversely associated with adherence to MD. For mild versus no decline, odds ratio (OR) comparing high to low MD adherence was 0.46 [95 % confidence interval (CI) 0.25–0.87, p = 0.012]. For substantial versus no decline, OR comparing high to low MD adherence was 0.34 (95 % CI 0.13– 0.89, p = 0.025). Among the nine MDS components, only vegetable consumption exhibited a significant inverse association with cognitive decline. Conclusions Closer adherence to the traditional MD is highly likely to protect against cognitive decline in this elderly Mediterranean population. Higher vegetable consumption appears to play a key role, possibly in synergy with additional components of the diet.
A. Trichopoulou · A. Kyrozis · M. Katsoulis · D. Trichopoulos Hellenic Health Foundation, 13 Kaisareias and Alexandroupoleos Street, 115 27 Athens, Greece
D. Trichopoulos · P. Lagiou Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA
A. Trichopoulou · D. Trichopoulos · P. Lagiou Bureau of Epidemiologic Research, Academy of Athens, 23 Alexandroupoleos Street, 115 27 Athens, Greece
C. La Vecchia Department of Clinical Sciences and Community Health, University of Milan, Via Venezian 5, 20133 Milan, Italy
A. Kyrozis (*) 1st Department of Neurology, Eginition Hospital, University of Athens Medical School, 74 Vas. Sofias Avenue, 11528 Athens, Greece e-mail: [email protected]; akyrozis@hhf‑greece.gr; akyrozis@ med.uoa.gr
P. Lagiou Department of Hygiene, Epidemiology and Medical Statistics, University of Athens Medical School, 75 M. Asias Street, Goudi, 115 27 Athens, Greece
M. Rossi · C. La Vecchia Dipartimento di Epidemiologia, IRCCS - Istituto di Ricerche Farmacologiche Mario Negri, via La Masa, 19, 20156 Milan, Italy
13
Keywords Mediterranean diet · Cognitive function · Elderly · Cognition · Dementia Abbreviations MDS Mediterranean diet score MMSE Mini-mental state examination
Introduction Studies of dietary patterns have gained prominence because of the accumulating evidence that combinations of foods and nutrients into certain patterns may act synergistically to provide stronger health effects than those conferred by their individual dietary components [1]. One of the best known such dietary patterns is the traditional Mediterranean diet (MD). It refers to a dietary pattern prevailing in Mediterranean countries and characterized by high consumption of fruits, vegetables, legumes and cereals’, monounsaturated fatty acids (with olive oil as the main added lipid), moderate consumption of alcohol (mainly wine and during meals) and low consumption of red meat and dairy products [2]. Closer adherence to MD has been inversely associated with the risk of cardiovascular diseases, cancer and overall mortality [3, 4]. There is also evidence that MD may protect against cognitive decline and dementia; recent reviews indicate that the accumulating evidence is supportive of such an effect, although there are inconsistencies in the reported data, as well as questions concerning the key component(s) of MD that may contribute to its apparent beneficial effect [5–9]. In the context of the population-based Greek segment of the European Prospective Investigation into Cancer and nutrition (EPIC-Greece), we studied the association of adherence to the traditional MD with change in cognitive function over a period of about 7 years. Our study was conducted among the elderly, in a population a large part of which still adheres to the traditional MD.
Materials and methods Subjects The European Prospective Investigation into Cancer and Nutrition (EPIC) is a prospective cohort study conducted in 23 centers in 10 European countries. EPIC aims to elucidate the role of biological, dietary, lifestyle and environmental factors in the etiology of chronic diseases. Its original focus was on cancer, but has been expanded to include several other diseases, including neurodegenerative conditions [10, 11]. EPIC was approved by the Ethical Review Board of the International Agency for Research on Cancer
13
Eur J Nutr
and by the local Ethics Committees in the participating centers. Participants gave informed consent before enrollment. Procedures were in line with the Helsinki declaration. The EPIC-Greece cohort consists of 28,572 volunteers, women and men, who were enrolled between 1994 and 1999 while aged between 20 and 86 years [3]. In EPICGreece, active follow-up evaluations are conducted every 3–5 years. At enrollment, an interviewer administered, validated semiquantitative food frequency questionnaire, including approximately 150 foods and beverages commonly consumed in Greece, was used to assess dietary intake [12]. Nutrient intakes were calculated with the use of a food composition database that has been modified to accommodate the special characteristics of the Greek diet [13]. A gradient of adherence to the traditional Greek MD was constructed using nine dietary components. A value of 0 or 1 was assigned to each of the nine components, using as cutoff values the sex-specific medians of consumption among the elderly participants (60 plus year of age) at time of enrollment and dietary assessment (baseline). For each of the five frequently consumed components in the context of MD (vegetables, legumes, fruit and nuts, cereals and fish), subjects were assigned a value of 0 if consumption was below the median and a value of 1 otherwise. In contrast, for each of the two less frequently consumed components in the context of MD (meat and meat products, dairy products), subjects were assigned a value of 1 if consumption was below the median and a value of 0 otherwise. For ethanol, a value of 1 was given to men who consumed quantities of ethanol from ≥10 to ≤50 g/day, whereas for women, the corresponding interval was from ≥5 to ≤25 g/day. For lipid intake, the ratio of monounsaturated to saturated fatty acids (MUFA/SFA) was used; those at or above the median scored 1 and those below the median scored 0. Thus, a MD score was constructed that could take a value from 0 (minimal adherence) to 9 (maximal adherence) [3]. Baseline socioeconomic, medical and lifestyle characteristics, notably age, years of schooling, marital status, physical activity and tobacco smoking were recorded with the help of trained interviewers. The frequency and duration of participation in occupational and leisure time physical activities were recorded [14] and allowed the calculation of a metabolic equivalent index (MET value) for each activity (household, professional, sporting and other activities) [15] and eventually an overall MET-hour sum, which indicates the amount of energy per kilogram of body weight expended during an average day by each participant. Tobacco daily quantity for current smokers was calculated as the number of cigarettes plus the number of cigars multiplied by 3 plus the number of pipes multiplied by 3. The sum is denoted as “tobacco quantity equivalent to cigarettes per day.” Standard measuring procedures were also used to
Eur J Nutr
assess anthropometric characteristics, with subjects wearing light clothing, no restrictive underwear and no shoes. Body weight was measured to the nearest 100 g, and height was measured to the nearest 1 cm. Body mass index (BMI) was then calculated as the ratio of weight over the square of height (in kg/m2). Cognitive function assessment For logistic reasons and to increase power, EPIC participants were considered eligible for the cognitive study if at the time of the planned first cognitive assessment, they were aged 65 years or older and resided in Athens or the surrounding Attica area. Using data from EPIC recruitment, 1,225 potential candidates were identified. However, among the 1,225 subjects, 480 were excluded due to the following reasons: death (151 subjects), moving outside the Attica region/not traced/refusal to participate (258). Therefore, 816 individuals remained for the first assessment of cognitive function, which was performed from May 2004 to October 2006. A second assessment of cognitive function took place from November 2011 to December 2012 in 401 participants among the 816 who had undergone the first assessment. Time intervals between the first (2004–2006) and the second (2011–2012) MMSE assessments ranged from 5.1 to 8.2 years (average 6.6 years, median 6.8 years). Exclusions from the second assessment were due to the following reasons: death (87 subjects) and missing data on baseline parameters or vital follow-up/moving outside the Attica region/not traced/refusal to participate (328 subjects) (Fig. 1). For the assessment of cognitive status, we used the mini-mental state examination (MMSE, 0 = minimum, 30 = maximum performance) [16], which has been validated in the Greek language [17]. MMSE was administered by health professionals (psychologists or “health visitors”) trained in the task and coordinated by author A.K., who is a neurologist. Three professionals administered and rated the first MMSE and three separate professionals the second one. There were no significant inter-rater differences of total MMSE score in either the first or the second administration. MMSE change between the two evaluations (cMMSE estimated as score in second minus score in first evaluation) was considered in three categories: improved or unchanged score (cMMSE ≥ 0), mildly lower score (cMMSE −1 to −4) and substantially lower score (cMMSE ≤ −5). Statistical analysis Frequency distributions were used for descriptive purposes. Median and interquartile range were computed for
Fig. 1 Flowchart: Baseline, first and second cognitive assessments
daily dietary intakes of MDS components, as well as total energy. In order to estimate the relation between adherence to MD and a change in MMSE (cMMSE) between −1 and −4 (slightly lower performance) and below −4 (substantially lower performance) versus a change ≥0 (no decline), we used multinomial logistic regression. Models were adjusted for sex, age (12 years, categorically), BMI (
ORIGINAL CONTRIBUTION
Mediterranean diet and cognitive decline over time in an elderly Mediterranean population Antonia Trichopoulou · Andreas Kyrozis · Marta Rossi · Michalis Katsoulis · Dimitrios Trichopoulos · Carlo La Vecchia · Pagona Lagiou
Received: 3 June 2014 / Accepted: 2 December 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Purpose Evidence suggests that dietary patterns compatible with the traditional Mediterranean diet (MD) may protect against cognitive decline. We prospectively assessed whether adherence to MD in the Mediterranean country of Greece is inversely associated with cognitive decline in the elderly and whether any particular MD component may play a key role. Methods Elderly men and women (N = 401) residing in the greater Athens area had dietary variables ascertained in 1994–1999. Adherence to MD was represented by the MD score [MDS, 0–3 (low), 4–5 (intermediate), 6–9 (high)]. The mini-mental state examination (MMSE) was administered by trained professionals to individuals aged 65 years or older in 2004–2006 (first assessment) and Electronic supplementary material The online version of this article (doi:10.1007/s00394-014-0811-z) contains supplementary material, which is available to authorized users.
re-administered in 2011–2012 (second assessment). MMSE change (cMMSE) was categorized as: improved/unchanged (cMMSE ≥ 0), mildly lower (cMMSE −1 to −4) or substantially lower (cMMSE ≤ −5). Associations were evaluated through multinomial logistic regression. Results Decline in MMSE performance was inversely associated with adherence to MD. For mild versus no decline, odds ratio (OR) comparing high to low MD adherence was 0.46 [95 % confidence interval (CI) 0.25–0.87, p = 0.012]. For substantial versus no decline, OR comparing high to low MD adherence was 0.34 (95 % CI 0.13– 0.89, p = 0.025). Among the nine MDS components, only vegetable consumption exhibited a significant inverse association with cognitive decline. Conclusions Closer adherence to the traditional MD is highly likely to protect against cognitive decline in this elderly Mediterranean population. Higher vegetable consumption appears to play a key role, possibly in synergy with additional components of the diet.
A. Trichopoulou · A. Kyrozis · M. Katsoulis · D. Trichopoulos Hellenic Health Foundation, 13 Kaisareias and Alexandroupoleos Street, 115 27 Athens, Greece
D. Trichopoulos · P. Lagiou Department of Epidemiology, Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA
A. Trichopoulou · D. Trichopoulos · P. Lagiou Bureau of Epidemiologic Research, Academy of Athens, 23 Alexandroupoleos Street, 115 27 Athens, Greece
C. La Vecchia Department of Clinical Sciences and Community Health, University of Milan, Via Venezian 5, 20133 Milan, Italy
A. Kyrozis (*) 1st Department of Neurology, Eginition Hospital, University of Athens Medical School, 74 Vas. Sofias Avenue, 11528 Athens, Greece e-mail: [email protected]; akyrozis@hhf‑greece.gr; akyrozis@ med.uoa.gr
P. Lagiou Department of Hygiene, Epidemiology and Medical Statistics, University of Athens Medical School, 75 M. Asias Street, Goudi, 115 27 Athens, Greece
M. Rossi · C. La Vecchia Dipartimento di Epidemiologia, IRCCS - Istituto di Ricerche Farmacologiche Mario Negri, via La Masa, 19, 20156 Milan, Italy
13
Keywords Mediterranean diet · Cognitive function · Elderly · Cognition · Dementia Abbreviations MDS Mediterranean diet score MMSE Mini-mental state examination
Introduction Studies of dietary patterns have gained prominence because of the accumulating evidence that combinations of foods and nutrients into certain patterns may act synergistically to provide stronger health effects than those conferred by their individual dietary components [1]. One of the best known such dietary patterns is the traditional Mediterranean diet (MD). It refers to a dietary pattern prevailing in Mediterranean countries and characterized by high consumption of fruits, vegetables, legumes and cereals’, monounsaturated fatty acids (with olive oil as the main added lipid), moderate consumption of alcohol (mainly wine and during meals) and low consumption of red meat and dairy products [2]. Closer adherence to MD has been inversely associated with the risk of cardiovascular diseases, cancer and overall mortality [3, 4]. There is also evidence that MD may protect against cognitive decline and dementia; recent reviews indicate that the accumulating evidence is supportive of such an effect, although there are inconsistencies in the reported data, as well as questions concerning the key component(s) of MD that may contribute to its apparent beneficial effect [5–9]. In the context of the population-based Greek segment of the European Prospective Investigation into Cancer and nutrition (EPIC-Greece), we studied the association of adherence to the traditional MD with change in cognitive function over a period of about 7 years. Our study was conducted among the elderly, in a population a large part of which still adheres to the traditional MD.
Materials and methods Subjects The European Prospective Investigation into Cancer and Nutrition (EPIC) is a prospective cohort study conducted in 23 centers in 10 European countries. EPIC aims to elucidate the role of biological, dietary, lifestyle and environmental factors in the etiology of chronic diseases. Its original focus was on cancer, but has been expanded to include several other diseases, including neurodegenerative conditions [10, 11]. EPIC was approved by the Ethical Review Board of the International Agency for Research on Cancer
13
Eur J Nutr
and by the local Ethics Committees in the participating centers. Participants gave informed consent before enrollment. Procedures were in line with the Helsinki declaration. The EPIC-Greece cohort consists of 28,572 volunteers, women and men, who were enrolled between 1994 and 1999 while aged between 20 and 86 years [3]. In EPICGreece, active follow-up evaluations are conducted every 3–5 years. At enrollment, an interviewer administered, validated semiquantitative food frequency questionnaire, including approximately 150 foods and beverages commonly consumed in Greece, was used to assess dietary intake [12]. Nutrient intakes were calculated with the use of a food composition database that has been modified to accommodate the special characteristics of the Greek diet [13]. A gradient of adherence to the traditional Greek MD was constructed using nine dietary components. A value of 0 or 1 was assigned to each of the nine components, using as cutoff values the sex-specific medians of consumption among the elderly participants (60 plus year of age) at time of enrollment and dietary assessment (baseline). For each of the five frequently consumed components in the context of MD (vegetables, legumes, fruit and nuts, cereals and fish), subjects were assigned a value of 0 if consumption was below the median and a value of 1 otherwise. In contrast, for each of the two less frequently consumed components in the context of MD (meat and meat products, dairy products), subjects were assigned a value of 1 if consumption was below the median and a value of 0 otherwise. For ethanol, a value of 1 was given to men who consumed quantities of ethanol from ≥10 to ≤50 g/day, whereas for women, the corresponding interval was from ≥5 to ≤25 g/day. For lipid intake, the ratio of monounsaturated to saturated fatty acids (MUFA/SFA) was used; those at or above the median scored 1 and those below the median scored 0. Thus, a MD score was constructed that could take a value from 0 (minimal adherence) to 9 (maximal adherence) [3]. Baseline socioeconomic, medical and lifestyle characteristics, notably age, years of schooling, marital status, physical activity and tobacco smoking were recorded with the help of trained interviewers. The frequency and duration of participation in occupational and leisure time physical activities were recorded [14] and allowed the calculation of a metabolic equivalent index (MET value) for each activity (household, professional, sporting and other activities) [15] and eventually an overall MET-hour sum, which indicates the amount of energy per kilogram of body weight expended during an average day by each participant. Tobacco daily quantity for current smokers was calculated as the number of cigarettes plus the number of cigars multiplied by 3 plus the number of pipes multiplied by 3. The sum is denoted as “tobacco quantity equivalent to cigarettes per day.” Standard measuring procedures were also used to
Eur J Nutr
assess anthropometric characteristics, with subjects wearing light clothing, no restrictive underwear and no shoes. Body weight was measured to the nearest 100 g, and height was measured to the nearest 1 cm. Body mass index (BMI) was then calculated as the ratio of weight over the square of height (in kg/m2). Cognitive function assessment For logistic reasons and to increase power, EPIC participants were considered eligible for the cognitive study if at the time of the planned first cognitive assessment, they were aged 65 years or older and resided in Athens or the surrounding Attica area. Using data from EPIC recruitment, 1,225 potential candidates were identified. However, among the 1,225 subjects, 480 were excluded due to the following reasons: death (151 subjects), moving outside the Attica region/not traced/refusal to participate (258). Therefore, 816 individuals remained for the first assessment of cognitive function, which was performed from May 2004 to October 2006. A second assessment of cognitive function took place from November 2011 to December 2012 in 401 participants among the 816 who had undergone the first assessment. Time intervals between the first (2004–2006) and the second (2011–2012) MMSE assessments ranged from 5.1 to 8.2 years (average 6.6 years, median 6.8 years). Exclusions from the second assessment were due to the following reasons: death (87 subjects) and missing data on baseline parameters or vital follow-up/moving outside the Attica region/not traced/refusal to participate (328 subjects) (Fig. 1). For the assessment of cognitive status, we used the mini-mental state examination (MMSE, 0 = minimum, 30 = maximum performance) [16], which has been validated in the Greek language [17]. MMSE was administered by health professionals (psychologists or “health visitors”) trained in the task and coordinated by author A.K., who is a neurologist. Three professionals administered and rated the first MMSE and three separate professionals the second one. There were no significant inter-rater differences of total MMSE score in either the first or the second administration. MMSE change between the two evaluations (cMMSE estimated as score in second minus score in first evaluation) was considered in three categories: improved or unchanged score (cMMSE ≥ 0), mildly lower score (cMMSE −1 to −4) and substantially lower score (cMMSE ≤ −5). Statistical analysis Frequency distributions were used for descriptive purposes. Median and interquartile range were computed for
Fig. 1 Flowchart: Baseline, first and second cognitive assessments
daily dietary intakes of MDS components, as well as total energy. In order to estimate the relation between adherence to MD and a change in MMSE (cMMSE) between −1 and −4 (slightly lower performance) and below −4 (substantially lower performance) versus a change ≥0 (no decline), we used multinomial logistic regression. Models were adjusted for sex, age (12 years, categorically), BMI (
