BMI, life-style and psychological conditions in a sample of elderly italian men and women

Abstract

Objectives

To identify the relationship among cognitive status, psychological conditions, anthropometric measurements and life-style in a sample of elderly Italian men and women.

Methods

Three hundred and six volunteers for ZINCAGE Project recruited. The sample was made up of healthy older adults living in the Marche Region aged 65 and over. All elderly were given a complete medical, anthropometric assessment, and psycho-social evaluation.

Results

Overall, the participants perceived themselves to be in very good or good (22%) or fair (69%) health; only 9% reported a poor health status. The 46% of the sample fell within the normal body mass index (BMI) range, though 38% were overweight, 12% were obese, and only 4% were underweight. In both sexes, BMI significantly decreased with age (p<0.001). BMI was positively associated with performing sedentary activities (r = 0.188; p < 0.001). Levels of both sedentary (r = 0.221; p < 0.001) and non-sedentary (r = 0.258; p < 0.001) leisure activities were positively associated with education level (p<0.05). It was found that lower scores of physical activity were associated to higher scores of Geriatric Depression Scale (r=−0.425; p<0.01), lower scores of Mini Mental State Examination (r=0.266; p<0.001) and higher score of Perceived Stress Scale (r=−0.131; p<0.05).

Conclusion

Men and women lead different lifestyles and have a different psychological status, with advancing age consequently stressing the need for healthy lifestyle programmes particularly in the case of overweight and obese elderly people.

Introduction

In today’s rapidly ageing society, policies for health and well-being throughout the life course are essential in promoting awareness and the highest possible level of health, especially in developing countries (1). Lifestyle can play a key role in influencing health in old age (2). Though the lifestyle characteristics in Italy and other European countries are abundantly described and documented (3, 4) there is a paucity of comparable international research in the literature (5, 6).

Older persons benefit substantially from regular physical exercise as this enables them to extend their independence by years (7). The amount of evidence in the literature supporting this is an unequivocal statement that physical activity assessment is a useful predictor of independence and disability (3). Recently, some findings have shown that regular exercise contributes to the overall promotion of good health and might also be an important recommendation for the prevention of many of the chronic and characteristic age-associated diseases, including diabetes mellitus and cognitive dysfunction (8).

In particular, a review of the literature reveals that men lead more active working and recreational lives than do women (9), who spend more time indoors doing household tasks experience an accompanying decline in physical activity (10).

Unfortunately, overall levels of physical exercise, recreational and social activity tend to fall with advancing age (11). This trend is more marked among the over 75s as they may underrate the substantial health benefits of physical activity (5). This circumstance can have health repercussions leading to overweight and obesity, whose increasing prevalence among the elderly has been widely publicised over the past few decades in both developed and developing countries (7, 12). Many factors and body composition changes are known to be associated with the ageing process and include a substantial reduction in free fat mass and an increase in adipose tissue (13). Surely, there is a scarcity of research for the assessment of sedentary life style in Italian elderly people. The identification of the parameters contributing to the causes of this is essential for devising prevention and management strategies (14). A mix of socio-economic, demographic and lifestyle factors as well as individual attitudes influence the risk of disease and obesity in the old (15). A number of European studies report findings in this area (16, 17). Some authors argue that poor dietary habits, nutritional disorders, physical inactivity as well as detrimental physiological, psychological, economic and social status are contributing factors to obesity (18). Indeed, evidence shows that there is a close relationship between BMI and cognitive changes which is linked to obesity (BMI ≥30 kg/m²) and the subsequent occurrence of clinical dementia, suggesting the involvement of BMI in brain function (19).

Given this background, the variables described seem to be related to health status in older adults (20). Because of the diversity in aspects, different lifestyle characteristics are to be expected among age groups and gender. So, the attention was focused on lifestyle factors and its potential relationship with overweight and obesity, respectively. Therefore, the findings obtained may be of significance when identifying the parameters contributing to the etiology of obesity and overweight in elderly Italians.

The main aims of this paper are to: 1) provide some descriptive data on lifestyle, health status and psychological aspects of the sample, identifying the differences of the latter variables between age classes and gender; 2) analyse the relationship between lifestyle, BMI and the psychological profile of a sample of elderly Italians.

Methods

Recruitment

The study was based on the parameters of the ZINCAGE project (contract n. FOOD-CT-2003-506850), a study of biochemical, genetic, psychological and lifestyle factors for healthy ageing (21). The project involved 17 groups belonging to 8 European countries. Old subjects aged 65 years and over were recruited from five different European countries (Italy, France, Germany, Poland, Greece), though the analyses reported here only apply to Italian older adults. As detailed described elsewhere (22), the study consisted of two parts: the first comprised the baseline assessment and the second consisted of a zinc supplementation trial. In the present paper we reported the results obtained at baseline.

The Italian sample consisted of 306 individuals. All participants lived in the Marche region, were healthy, non-institutionalised and did not require day care and/or nursing support volunteers. Older adults were selected according to the SENIEUR protocol (23). So, the participants were free of medication such as steroids, diuretics, anticonvulsants, antidepressive drugs, antibiotics. Subjects were also excluded if they had autoimmune, neurodegenerative, cardiovascular, kidney or liver diseases, diabetes, cancer, chronic inflammatory bowel disease, sickle cell anaemia and endocrine disorders. Ethical approval was obtained from the National Ethical Committee. All participants signed informed consent form before taking part in the study. The sample was divided into three age groups: 141 of the participants were 65-74yr, 101 were 75-84yr and 64 were 85 or above.

Instruments

All participants were assessed for psychological status, cognitive functions, lifestyle, prescribed medications, including the following.

Health status

Information about medical history, medications, surgery, illnesses, pains and fractures were collected using the “Health and Morbidity Questionnaire”. Detailed history was taken regarding present and past illnesses. Participants were also asked to rate their overall health as very good, good, fair, poor, very poor. The questionnaire used reflects that one used to determine the health status in elderly and centenarian subjects and validated in more than n. 500 old individuals (IMUSCE study) (24).

Lifestyle characteristics

The interviewer-administered “Lifestyle Questionnaire” was used to obtain information on lifestyle, socio-demographic and family characteristics (age, gender, education, marital status, kind of household), economic status, smoking and dietary habits, alcohol consumption, and physical activity. This instrument was used in accordance to ZENITH Project (3). It was developed and validated by the Northern Ireland Centre for Food & Health (NICHE), and translated into Italian. The questions were suitable for elderly population.

As regards smoking habits, participants were classified in three groups, such as 1) current smokers; 2) ex-smokers and 3) non-smokers. Alcohol consumption was assessed asking the type of drink (beer, white wine, red wine, liqueurs and spirits), frequency (daily, weekly or monthly) and quantity of alcohol intake.

Physical activity included indoor, outdoor and leisure activities. Frequency of physical activity, measured on a daily or weekly basis, and the type of activities (preparing food, cooking and washing up, doing the laundry and ironing, shopping, doing the housework) carried out were asked to participants. Moreover, both sedentary and non-sedentary leisure activities were assessed (25). Non-sedentary recreation included cycling, dancing, gardening, exercising, games (e.g. Italian “bocce”), and walking. Sedentary leisure activities included reading, watching TV or listening to the radio, playing cards or chess and knitting or sewing.

Two new variables were created for non-sedentary and sedentary activities, as the sum of the frequencies of each activity. Moreover, the number of leisure activities was calculated for each subject, with a range of 0-10.

Anthropometry

Height and body weight of participants were recorded. Height was measured in the standing position with bare foot against the wall. Body weight was measured in kilograms using a scale to the nearest 0.5 kg with light clothes on. BMI was calculated as weight in kilograms divided by the square of the height in metres (kg/m²), and the World Health Organization classification of BMI (26) for adult population was applied: underweight, BMI < 18.5 kg/m²; normal weight, 18.5 ≤ BMI < 25.0 kg/m²; overweight, 25.0 ≤ BMI < 30.0 kg/m²; obesity, BMI ≥ 30.0 kg/m².

Psychological assessment

Evaluation of global cognitive abilities was performed by means of Folstein’s Mini Mental State Examination (MMSE). This instrument is reliable, valid and suitable for elderly people (27). It is translated and validated in many countries, including Italy (28). It has been used as a research tool to screen for cognitive disorders in epidemiological studies and follow cognitive changes in clinical trials. While the MMSE has limited specificity with respect to individual clinical syndromes, it represents a brief, standardized method by which to grade cognitive mental status. A cut off score of 23 out of 30 indicates possible cognitive impairment. Both age and education affect the rating which was adjusted for the two variables using the Italian correction table (29).

The Geriatric Depression Scale (GDS-15; 30) was used to screen for any elements of depression. This scale was developed to assess many of the problems associated with depression and for identifying depressive symptoms in older adults. It has been used in a variety of settings and with appropriate age groups 55-100 years old. Total scores were obtained by adding one point for each response which was symptomatic of depression, giving a score range of 0-15, where a score ≥ 5 (cut off value) suggested possible depression. It has proven reliability and validity. The Italian version was used (31).

As its name suggests, the Perceived Stress Scale (PSS; 32) was used to evaluate perceived stress. This instrument is the most widely used psychological instrument for measuring the perception of stress, it is a measure of the degree to which situations in one’s life are appraised as stressful. It has substantial reliability and validity and it is validated in Italian subjects (33). It asks how often a person has felt or thought a certain way over the last month. Items were designed to tap how unpredictable, uncontrollable, and overloaded respondents find their lives. The scale also includes a number of direct queries about current levels of experienced stress. Each item is rated on a 5-point scale, which ranges from 0 to 56. A score greater than the mean value indicated a high level of stress. Cronbach’s Alpha for GDS and PSS tests in our sample was 0.74 and 0.86 respectively.

Statistical Analysis

Analyses were carried out using SPSS 16.0 statistical software for Windows (SPSS; Chicago, IL, 2002). A descriptive analysis was firstly performed, to evaluate the distribution of the variables in the sample, in terms of gender (male, female) and age groups (65-74, 75-84, 85+). The results were expressed as mean (SD) or percentages as appropriate. Comparisons of mean values between groups were made by two independent samples t-tests or one-way ANOVA followed by post-hoc test as appropriate, while percentages were compared using chi-square test. The second step performed was done by Pearson’s correlation analysis to assess associations between the variables above mentioned and the BMI and psychological tests scores in the total sample. Finally, comparisons of percentages of the latter variables (psychological tests or BMI) were evaluated with chi-square test. The sample was divided in 2 or more groups on the basis of the variable considered, as described in the section “Instruments”. A value of p<0.05 was accepted as statistically significant.

Results

Socio-demographic variables

A descriptive analysis of the characteristics of the ZINCAGE Italian sample, by age and gender, is provided in Table 1. The mean age of the total sample was 76.91 (8.5) years. Distribution of age between the two sexes was not statistically different (p>0.05).

Table 1.

Characteristics of the sample

	Age group				Gender			Total
	65-74	75-84	85+	p-Value	Female	Male	p-Value
	n = 141	n = 101	n = 64		n = 173	n = 133		n=306
Marital status				<0.001			<0.001
Married/full time relationship	78%	48%	22%		40%	77%		56%
Single	5%	3%	8%		4%	7%		5%
Widowed	17%	49%	70%		56%	16%		39%
Mean years of education (SD)	5.02	4.39	2.83		3.84	5.02		4.35
	(2.3)	(2.6)	(1.6)	<0.001	(2.3)	(2.5)	<0.001	(2.4)
Mean BMI (SD)	26.2	24.8	23.3		24.6	25.9		25.19
	(3.6)	(3.4)	(4.1)	<0.001	(4.0)	(3.5)	<0.01	(3.8)
Smoking habits				<0.001			<0.001
Smokers	6%	4%	0%		3%	5%		4%
Ex-smokers	35%	27%	20%		3%	65%		30%
Non smokes	59%	69%	80%		94%	30%		66%
Alchool consumption				n.s.			<0.001
Everyday	67%	71%	62%		59%	78%		68%
Occasionally / sometimes	6%	6%	3%		5%	6%		5%
Never	26%	23%	35%		36%	16%		27%
MMSE				<0.001			n.s.
Possible cognitive impairment	32%	35%	62%		43%	33%		39%
Not cognitive impairment	68%	65%	38%		57%	67%		61%
GDS				<0.001			<0.001
Possible depression	24%	40%	56%		46%	22%		36%
Not depression	76%	60%	44%		54%	78%		64%
PSS				n.s.			<0.001
Stressed	6%	5%	9%		11%	0%		6%
Not stressed	94%	95%	91%		89%	100%		94%
Mean number of leisure activities (SD)	4.20	3.65	2.44		3.65	3.64	n.s.	3.65
	(1.2)	(1.2)	(1.3)	<0.001	(1.5)	(1.2)		(1.4)
Level of sedentary activities (SD)	5.64	5.48	4.53		5.49	5.19		5.36
	(2.1)	(2.2)	(2.1)	<0.001	(2.3)	(2.1)	n.s.	(2.2)
Level of non-sedentary activities (SD)	4.55	2.87	1.13		2.72	4.0		3.28
	(2.5)	(2.1)	(1.4)	<0.001	(2.2)	(2.7)	<0.001	(2.5)

Note: Percentages compared with chi-square test; mean (SD) compared with t-test (gender) or ANOVA (age groups).

Fifty-six percent of the participants were married, 39% reported that they were widowed and 5% had never been married or were currently single. The marital status was significantly different between gender and age groups (p<0.001).

As regards gender, the level of education was higher in male respect to female (p<0.001). Referring to age groups, the level of education was higher in the 65-74 age group respect to the 75-84 and 85+ groups (p<0.001).

Health status

Table 2 shows the health status characteristics of the sample. The majority of the participants perceived themselves to be in very good or good health (22%) or fair (69%); only 9% reported a poor health status. Men (24%) considered their health status significantly better than did women (20%) (p<0.05). A high percentage of the sample (92%) was taking medications. The mean number of prescribed drugs was the same for men and women, such as 2.55 (2.23) and 2.56 (1.88) respectively. The most common diseases or chronic conditions were hypertension (48%), cataracts (24%), arrhythmia (20%), heart failure (19%), osteoporosis (17%), gastric ulcer (11%), chronic bronchitis (11%), allergies (9%), and migraine (6%). The mean number of diseases for women was significantly higher (3.64 (1.83)) than that of men (3.02 (1.81)) (p<0.01).

Table 2.

Health status characteristics

	Female n = 173	Male n = 133	Total n=306	p-Value
Self-reported health status				<0.05
Very good/good	20%	24%	22%
Fair	67%	71%	69%
Poor/very poor	13%	5%	9%
Mean number of prescribed	2.56 (1.88)	2.55 (2.23)	2.56(2.04)	n.s.
drugs (SD)
Mean number of diseases	3.64 (1.83)	3.02 (1.81)	3.38 (1.84)	<0.05
(SD)
Chronic conditions or diseases
Hypertension	55%	41%	48%
Cataract	22%	27%	24%
Arrhythmia	18%	23%	20%
Heart failure	19%	19%	19%
Osteoporosis	26%	5%	17%
Gastric ulcer	10%	14%	11%
Chronic bronchitis	10%	12%	11%
Allergy	11%	8%	9%
Migraine	9%	4%	6%

Note: Percentages compared with chi-square test; mean (SD) compared with t-test.

Cognition and psychological assessment

Table 1 lists the psychological aspects of the sample by age and gender. Possible cognitive impairment as evidenced by a MMSE score below 24/30 was detected in 39% of the sample. The prevalence of possible cognitive decline was higher in older subjects (p<0.001). As regards gender, percentage of possible cognitive decline did not differ between male and female (p>0.05); only the mean MMSE score value was significant, such as 24.73 (3.49) for men and 23.58 (4.14) for women (p<0.01). As highlighted in Table 1, the results showed not significant differences between gender in possible cognitive decline (p>0.05). Thirty-six per cent of the sample had possible depression, assessed by means of the cut-off value of GDS ≥ 5 (30). The mean score value of the sample was 2.90 (2.34) for men and 4.52 (2.90) for women (p<0.001). Mild depression, defined with a GDS score of 6-10, was observed in 32% of the sample, whereas severe depression, defined with a score of >10, was observed in 3%. A low percentage of the sample perceived to have a high level of stress (6%). As measured by PSS, this result represents a score greater than the mean value of the scale. However, the women were significantly more susceptible to stress than the men (p<0.001), with a mean PSS value of 16.7 (8.40).

The correlation analysis showed that depression (r=0.346; p<0.001) and cognitive status (r=-0.361; p<0.001) became significantly worse with advancing age. Stress level was not associated with age (p>0.05).

Table 3 shows depressive symptoms, cognitive and stress status by some socio-demographic characteristics. The majority of subjects with possible cognitive decline (76%), possible depression (64%) and stress (67%) had no education. As regards marital status, the majority of subjects with no cognitive impairment or indicators of depression (67%) were married or engaged in full time relationship. The prevalence of underweight, normal weight and obesity did not differ significantly between subjects grouped on the basis of depressive symptoms, cognitive and stress status (p>0.05).

Table 3.

Depressive symptoms, cognitive and stress status by some socio-demographic characteristics

	MMSE (%)			GDS (%)			PSS (%)
	Possible cognitive impairment	Not cognitive impairment	p-Value	Possible depression	Not depression	p-Value	Stressed	Not stressed	p-Value
Level of education			<0.001			<0.001			<0.01
No education	76%	32%		64%	40%		67%	47%
Low education	23%	53%		30%	48%		11%	44%
Middle/high education	1%	15%		6%	12%		22%	9%
Kind of household			<0.001			<0.001			n.s.
Alone	15%	15%		20%	12%		28%	14%
With partner/partner and children	40%	66%		36%	67%		50%	57%
With other persons	45%	19%		44%	21%		22%	29%
Marital status			<0.001			<0.001			n.s.
Married/full time relationship	40%	67%		36%	67%		50%	57%
Single	4%	6%		6%	5%		11%	5%
Widowed	56%	27%		58%	28%		39%	38%
BMI			n.s.				n.s.		n.s.
Underweight	5%	3%		4%	3%		7%	3%
Normal weight	49%	44%		52%	44%		33%	47%
Overweight/obesity	46%	53%		44%	53%		60%	50%
Mean number of leisure activities (SD)	3.26	3.98	<0.001	3.08	4.03	<0.001	2.89	3.76	<0.01
	(1.3)	(1.3)		(1.5)	(1.2)		(1.4)	(1.3)
Level of sedentary activities (SD)	4.96	5.72	<0.001	4.80	5.75	<0.001	4.05	5.51	<0.01
	(1.7)	(2.2)		(1.9)	(2.1)		(1.4)	(2.1)
Level of non-sedentary activities (SD)	2.81	3.69	<0.001	2.14	4.0	<0.001	1.78	3.46	<0.01
	(2.6)	(2.4)		(2.1)	(2.5)		(1.8)	(2.5)

Note: Percentages compared with chi-square test; mean (SD) compared with t-test.

BMI

Mean weight was 73.05 (11.62) kg in men and 61.12 (10.75) kg in women, mean height was 167.6 (6.0) cm in men and 157.4 (4.7) cm in women. These measurements were significantly different between age groups (p<0.001) and gender (p<0.01), as showed in Table 1. Mean BMI was 25.9 (3.5) kg/m² in men and 24.6 (4.0) kg/m² in women (p<0.01). So, in terms of the mean BMI value, men were overweight and women were normal weight. Overall, 4% of the sample was underweight (1% of men and 5% of women), 46% were within the normal BMI range (41% of men and 51% of women), 38% were overweight (42% of men, 35% of women), and 12% were obese (16% of men and 9% of women).

The correlation analysis showed that BMI is negatively associated with age (r=-0.302; p<0.001).

Table 4 shows BMI by some socio-demographic characteristics. There were no significant differences between BMI categories and educational level. There were some significant differences in the three categories of BMI in accordance to kind of household and marital status (p<0.05). The majority of overweight/obese subjects was married (67%), while the majority of underweight (60%) was widowed (p<0.05).

Table 4.

BMI by some socio-demographic characteristics

	BMI (%)
	Underweight	Normal weight	Overweight/obesity	p-Value
Level of education				n.s.
No education	80%	52%	43%
Low education	10%	40%	46%
Middle/high education	10%	8%	11%
Kind of household				<0.05
Alone	40%	16%	11%
With partner/partner and children	30%	51%	66%
With other persons	30%	33%	23%
Marital status				<0.05
Married/full time relationship	30%	51%	67%
Single	10%	5%	3%
Widowed	60%	44%	30%

Note: Percentages compared with chi-square test.

Smoking habits and alcohol consumption

The number of smokers in the sample was generally low (4%), and the percentage of men was higher (5%) than that of women (3%), but not significantly different (Table 1). Thirty percent were classified as ex-smokers. Based on the level of alcohol consumption, 68% of our sample consumed alcohol everyday, in particular red and white wine and beer. Indeed, results showed that men tended to consume significantly more alcohol than women (p<0.001); in fact 78% of men in respect 59% of women consumed alcohol everyday.

The correlation analysis showed that there are no associations between alcohol consumption and age, BMI and psychological aspects (p>0.05).

Fifty percent of possible depressed people and 61% of older adults with possible cognitive decline drank alcohol everyday (p<0.05).

Physical and leisure activities

Participation rates in leisure activities were low for the whole sample, considering the mean number of leisure activities carried out by participants (Table 1). The latter was 3.65 (1.4) (range 0-10), with no significant differences between the sexes. Age negatively affected the number of leisure activities performed; indeed the mean value decreased from 4.20 (1.2) for subjects aged 65-74 to 3.65 (1.2) for those aged 75-84 and to 2.44 (1.3) for those aged 85 and above. Some leisure activities, such as reading, engaging in regular physical exercise, dancing and play cards were done more by younger than old age groups, where the level of p<0.05 was significant. However, women tended to perform more indoor/outdoor activities than men (p<0.001) who tended to perform more non-sedentary leisure activities than women (p<0.001).

The correlation analysis showed that overall levels of physical activity (PA) declined with increasing age (r = -0.535; p<0.001). Levels of both sedentary (r = 0.221; p<0.001) and non-sedentary (r = 0.258; p < 0.001) leisure activities were positively associated with education level. BMI was positively associated with performing sedentary activities (r = 0.188; p<0.001). Psychological status was also studied in relation to physical activity and it was found that lower scores of PA were associated to higher scores of GDS (r=-0.425; p<0.01), lower scores of MMSE (r=0.266; p<0.001) and higher score of PSS (r=-0.131; p<0.5).

Indeed, older adults who had received secondary education as well as those who went on to higher education tended to walk, garden and exercise more than participants with little or no education (data not shown; p<0.01). Also, being married or in a full time relationship tended to make subjects significantly more active with respect to single or widowed elderly individuals, as regards non-sedentary leisure activities (data not shown; p<0.001).

Table 3 shows the psychological conditions by levels of physical activity and number of leisure activities. People with depressive symptoms, possible cognitively impaired and stressed older adults showed lower levels of sedentary and not-sedentary leisure activities than those with no depression, cognitive decline or stress (p<0.001). The situation was the same if the mean number of leisure activities was considered (p<0.001).

Discussion

These findings represent an important step forward in clearly defining some of the characteristics essential to healthy ageing. The ZINCAGE project has contributed invaluable insight into the diversity among elderly Europeans. The main purpose of this paper is to describe the psychological status, anthropometric measurements, lifestyle characteristics and to determine the relationship among these variables in a sample of elderly Italians. Despite the indisputable importance of these factors and their inter-relationships, Italian research continues to lag behind. Firstly, it must be stressed that defining what is meant by “healthy elderly people” was not a simple task and even though older adults perceived themselves as healthy, upon objective investigation certain common age-related diseases were found to be prevalent among them, as previous findings confirm (34). In contrast, however, the majority of participants perceived themselves to be in very good or good health, with more favourable self-rated health status among men, yielding a significant difference.

An important finding of our study which emerged from the assessment of the prevalence of BMI-derived overweight and obesity is consistent with a very recent survey conducted in Italian adults, of whom over 30% were overweight and about 8% obese (35). We also compared these percentages with other results from previous ZINCAGE and European survey papers (14, 36). We have documented the different distribution of socio-demographic, behavioural and environmental factors by BMI. Among them, we found that specific indicators of social factors, such as marital status and kind of household are responsible for the link with BMI, as was also concluded by Inelmen et al. (2008) (35). Indeed, we found that BMI was inversely and significantly associated with age. Accordingly, the mean values of anthropometric measurements significantly differed among age groups and the sexes, as is consistent with similar studies (37, 38). The BMI of Italian elderly people is favourable and is probably due to healthy dietary habits (39).

Analysis of lifestyle and dietary habits carried out within the scope of this study revealed that 68% of our sample consumed alcohol everyday and that wine was the preferred beverage. This is a characteristic feature of Italians. Furthermore, based the WHO’s per capita alcohol consumption figures alone from its Global Status Report on Alcohol, Italy ranks as an average consumer of alcohol (40). An analysis of the actual patterns of alcohol consumption and demographics showed that men tended to consume significantly more alcohol than women and that alcohol consumption fell significantly in the older age groups.

Regarding physical activity, we assessed both sedentary and non-sedentary physical activities. Our findings reveal that elderly Italians partake in low levels of PA. This result is supported by the ILSA study which estimates that 50% of the total burden of diseases in men and 25% in women are dependent on unhealthy lifestyles (41).

Comparing with gender, it emerged that women engaged in less physically activity than men (42). Our study results and those of others confirm that while “walking” is one of the most common forms of physical exercise practised by older persons (43), it is watching TV that most occupies their leisure time (44). As regards socio-demographic aspects, the higher educational level was significantly associated with a higher level of active physical activity. Moreover the subjects married or in a full time relationship tended to be significantly more active with respect to single or widowed elderly individuals, as regards non-sedentary activities. These results are consistent with a very recent study by Skrzypczak et al. (2008) (37). In the present study also BMI was positively associated with performing sedentary activities, as also argued recently by Vaughan (2009) (45). Our analysis comprised also the study of psychological status in relation to PA. The positive influence of physical activity on psychological status in elderly is well known (46). We demonstrate that depressed and possible cognitively impaired older adults tended to perform some physical activities, as walking, significantly less than do other subjects. Indeed, several studies show that depression and other psychological symptoms are more common in people whose physical activity is low than in their active counterparts (8). In this context, the link of overweight and obesity with psychological status is very important, in accordance with several studies (47). As for whether there exists an association between BMI and depression, Roberts et al. (2003) (48) propounded four theories in the Alameda County Study. The first one is that obesity increases the risk of depression, whereas the second hypothesises that, vice versa, depression increases the risk of obesity. In the third case, there may be a reciprocal relationship between the two. Lastly, there is no association between them. Similarly, our study was unable to establish a difference in overweight/obesity prevalence between subjects with or without possible depression and stress. The authors argue that a number of explanations are conceivable, such as the possible role of biological, social and lifestyle factors (48, 49). In the same way, although other paper reports a significant relationship between increased BMI and cognitive performances in healthy elderly men and women (50), in our study subjects with and without possible cognitive decline had the same distribution in terms of overweight and obesity. Probably, this result may be due to a limited number of our sample. However, other factors can be involved in this relationship, such as the direct action of adiposity on neuronal tissue through neurochemical mediators produced by the adipocytes. For example, some authors support the role played by leptin, not only in the modulation of food intake, but also in relation to learning and memory processes (51). Since serum leptin was not herein performed, this role may be only hypothesised in our samples supposing possible increased levels of leptin in both groups (obese and normal weight). Such an hypothesis is supported by the discovery that higher serum leptin appears to protect against cognitive decline, independent of comorbidities and body fat in elderly individuals (52). However, further studies are in progress in a more wide number of samples.

Our study has some limitations too. In particular, it is important to point out that the ZINCAGE study recruited volunteers. Cognition screening of the sample represented neither the general population nor cognitively intact older adults.

In conclusion, although the results not suggest a relationship between BMI and psychological functions, further research will be needed to elucidate the mechanisms involved, such as the comorbidity (53). Moreover, the importance of monitoring BMI in older people emerged. In this study, only the data of the Italian sample were reported, because the knowledge about the factors related to BMI and lifestyle characteristics in Italian elder people is sparse. Further analysis are needed in order to analyse the differences among European countries. The identification of these may be of value for research on old people, in particular those who live alone and have a low educational level. Additionally, the relationship between physical activity and psychological function in healthy elderly people is of importance. As emerged, older women perform less physical activity than men, so, the role of involved nutritional and socio-demographic aspects should be investigated. As argued by Varo et al. (2003) (6), smoking and regular exercise are central factors in transforming an unhealthy lifestyle as both are potentially modifiable. Surely, lifestyle changes towards a healthier future for the elderly could have a major positive impact on quality of life, in terms of decreased disability and morbidity with advancing age (54).

Acknowledgments: The study was supported financially by the European Commission (ZINCAGE Project, contract n. FOOD-CT-2003-506850) and by INRCA.

Financial disclosure: None of the authors had any financial interest or support for this paper.