«Wild Edible Plant Consumption and Age-Related Cataracts in a Rural Lebanese Elderly Population: A Case control Study By Joelle Zeitouny School of ...»
The distribution of the average energy intakes per day over a period of three months by the participants (71% women) was divided according to their BMI and gender and is described in quantiles along with the mean (Table 4.3). The 90th percentiles values for both normal weight and overweight men and women were below the FAO/WHO/UNU recommended values.
Controls had a significantly higher intake of energy, lutein and zeaxanthin, βcarotene, vitamin A and vitamin C than cases (Mann-Whitney U test, p0.001) (Table 4.4). On the other hand, intake of α-tocopherol was not significantly different between cases and controls (Mann-Whitney U test, p0.3).
Average intake of lutein and zeaxanthin per day over a period of three months (μg) was significantly correlated with the intake of wild leafy greens (ρ = 0.332, p0.01) using Spearman’s rho correlation coefficient (as the distribution of both variables was markedly skewed). However, the major contributors of lutein and zeaxanthin did not include wild leafy greens (Table 4.5). The three main sources of lutein and zeaxanthin in the diet of subjects with a high intake of lutein and zeaxanthin (N=67) were in fact spinach, Swiss chard and parsley. The main sources of lutein and zeaxanthin did not differ between the diet of cases (N=26) and controls (N=41) that had a high intake of these two nutrients (Table 4.6).
More than 90% of both cases and controls consumed wild leafy greens.
Around 40% collected them and 90% of those learned about wild leafy greens and their identification and collection from their parents (Table 4.7).
The average intake of wild leafy greens (g) during their growing season (a period of three months) was very similar between cases and controls (Table 4.7).
Roughly 18% of the variation in intake could be explained by a simple model
equation that was derived from linear regression analysis (Table 4.8):
I = 0.15 K + 0.29 A + 0.18 U + 0.69 where I is the log-transformed intake (g) of wild leafy greens and K, A and U represent the following predictors: knowledge on wild leafy greens acquired from parents, agriculture as present or past occupation and unemployment, respectively.
4.4 DIVERSITY SCORES
Controls had a significantly higher Food Variety Score (FVS) than cases and consumed a significantly higher number of food items rich in lutein and zeaxanthin, β-carotene, vitamin C and α-tocopherol than cases did over a period of three months (Mann-Whitney U test, p0.001) (Table 4.9). However, the number of food items rich in vitamin A consumed did not significantly differ between cases and controls (Mann-Whitney U test, p0.5).
Likewise, there was no significant difference between the Dietary Diversity Scores of cases and controls neither when leafy greens (Mann-Whitney U test, p0.6) nor when wild leafy greens (Mann-Whitney U test, p0.06) were considered as part of the 8 groups we divided our food items into (Table 4.10).
Since cases and controls were matched according to age, there was no age difference between the two groups. Males and females as well as smokers and neversmokers seemed to have the same risk of having age-related cataracts whereas in the literature, females and smokers had a higher risk of contracting age-related cataracts than males and never-smokers respectively (Hennis et al., 2004; McCarty et al., 1999;
DeBlack et al., 2003). It is, however, important to note that recruited cases were mostly female (142 versus 58 males). Since recruitment was random, the high number of females might indicate a higher prevalence of the disease among Hermel females.
Even though dark-colored irises put the subjects at higher risk of contracting agerelated cataracts (Hammond et al., 2000), less cases had dark-colored irises than controls. On the other hand, more cases than controls have worked in agriculture, and have thus had higher exposure to the sun; in the literature, UV light exposure has been consistently found to be associated with age-related cataracts (McCarty & Taylor, 2002).
This study demonstrates, in agreement with the literature, that intakes of lutein and zeaxanthin, β-carotene, vitamin A and vitamin C are inversely related to agerelated cataracts (Mares-Perlman et al., 1995; Chasan-Taber et al., 1999; Lyle et al., 1999a; Taylor et al., 1991, 2002; Mares-Perlman et al., 2000; Jacques et al., 2001;
Chylack et al., 2002). Intake of α-tocopherol might possibly be related as well to agerelated cataracts (Vitale et al., 1993; Leske et al., 1998; Rouhiainen et al., 1996;
Mares-Perlman et al., 2000); however, the average intake values reported in this study for α-tocopherol were extremely low, most likely because the semi-quantitative food frequency questionnaire that was used did not include vegetable oil and nuts, which are the richest sources of α-tocopherol along with olive oil in the study population’s diet. In Lebanon, olive oil is most frequently consumed raw and rarely used in cooking because it is generally more expensive; the most common cooking fat is vegetable oil (Nasreddine et al., 2006). Around 18% of the study participants had a low intake of olive oil compared to the average intake of olive oil per capita per day 50 in Lebanon (FAOSTAT, 2003), probably due to under-reporting, which partly explains the low energy intake.
In fact, even though the food frequency questionnaire was not exhaustive, considerable under-reporting of energy intake is likely to have occurred as the 90th percentiles values for all study participants were below the FAO/WHO/UNU recommended average energy intake values. A number of factors are associated with low-energy reporting including weight status, age, sex effects, socioeconomic effects, health-related activities, behavioral effects, and psychological effects (Livingstone & Black, 2003). However, the single most consistent factor related to under-reporting is weight status: the probability that a subject will under-report generally increases as BMI increases (Briefel et al., 1997; Johansson et al., 1998) and as previously stated, 81% of the study participants had a BMI over 25 kg/m2. The Goldberg cut-off method, which evaluates self-reported energy intakes against estimated energy requirements, could not be applied since food frequency questionnaires are designed to represent a person’s usual eating habits over a period of time and are not a precise measure of energy intake. A 24-hour recall could have probably better described the diet in terms of energy intake (Gibson, 2005). Consequently, the quantiles of intake of lutein and zeaxanthin, β-carotene, vitamin A, vitamin C, and α-tocopherol are probably inaccurate; therefore the comparison among cases and controls is of interest rather than the absolute values themselves which need to be interpreted more cautiously.
Even though the average daily intake of lutein and zeaxanthin was significantly correlated with the intake of wild leafy greens (ρ = 0.332, p0.01), the major contributors of lutein and zeaxanthin in the diet of both cases and controls did not include wild leafy greens. This could be either due to high dietary diversity in both groups (see later) or to incorrect nutrient content information for wild leafy greens (as data were not available for most wild leafy greens and average nutrient content values had to be estimated from databases).
51 On the other hand, results showed that there was no difference in intake of wild leafy greens between cases and controls (although the variability in intake was very high). When the number of wild leafy greens consumed over the last three months was divided by the total number of food items consumed, a significant difference was observed between participants who classified themselves as “poor” and others as “middle-class” (Mann-Whitney U test, P0.04). In other words, participants who classified themselves as “poor” consumed more wild leafy greens than their “middle-class” counterparts regardless of whether they had age-related cataracts or not. Therefore, wild leafy greens are substantially contributing to the quality of the diet (of those who classify themselves as “poor”) through improving the micronutrient content of the latter (Humphry et al., 1993; Ogle et al., 2001).
Anecdotal evidence indicates that there is no taboo against the consumption of wild leafy greens. On the contrary, most people would consume wild leafy greens if the latter were readily available to them. As a matter of fact, over 90% of both cases and controls consumed wild leafy greens while only 40% collected them. In this regard, it is noteworthy to mention that our study participants were relatively old (average age is 66) and most likely relied on their offspring to supply them with wild leafy greens since it is common practice in the area where the study took place to collect wild edible plants in the nearby agricultural fields, gardens, and mountains, during their growth season. There is however no data on the number of participants who possessed knowledge of wild leafy greens’ identification and collection to show whether new generations were less likely to identify edible wild food resources plants that existed in their lands compared to older ones ( Ladio & Lozada, 2003).In fact, knowledge about wild plants, especially in poor rural areas, is often associated with better health status. In a study by McDade et al. (2007) conducted in the Bolivian Amazon, mothers with higher levels of plant knowledge and use had healthier children, independent of potentially confounding variables related to education, market participation, and acculturation.
52 Around 90% of those who collected wild leafy greens learned about their identification and collection from their parents and that was one of the strongest predictors of wild leafy greens’ intake. Cultural transmission from parent-to-child is called vertical, and vertical transmission is associated with slower change in knowledge systems and is thus protective of traditional knowledge (Cavalli-Sforza & Feldman, 1981). Vertical transmission also helps to maintain diversity of knowledge and beliefs within a population (Hewlett & Cavalli-Sforza, 1986). Therefore, traditional knowledge is probably still strong in the community where the study took place, and is ensuring continued use of wild plants.
Other predictors of wild leafy greens’ intake included agriculture (as past or present occupation) and unemployment. Both of these predictors are indicative of poor socio-economic status. In this study, participants were asked to rate their subjective perception of their socio-economic status and more than two-thirds of them classified themselves as poor. Asset-based assessments of socio-economic status, such as household survey and participatory wealth ranking (Hargreaves et al., 2007), could not be done because of time and financial constraints.
Nonetheless, the R-squared value of the derived model was of only 0.173, meaning that the above-mentioned variables were not exclusively able to predict the intake of wild leafy greens. Other variables, such as access for example (Swindale & Bilinsky, 2005), as well as random effects related to intake may partially account for such a low R-squared value. However, ethnicity and religion are unlikely to predict intake because, as stated before, the study was conducted in a homogeneous community.
In general, controls had a more diverse diet than cases even though they did not have a significantly higher BMI (Mann-Whitney U test, P0.5) - dietary diversity has been associated with BMI in some studies (Benefice et al., 2007) as access to more diverse foods can sometimes lead to diets higher in fats, causing weight gain and resulting in health problems (Drewnowski & Popkin, 1997). Controls consumed a 53 significantly higher number of food items rich in lutein and zeaxanthin, β-carotene, vitamin C and α-tocopherol than cases did over a period of three months (MannWhitney U test, p0.001). However, the number of food items rich in vitamin A consumed did not differ between cases and controls although the p value obtained was very close to being significant (Mann-Whitney U test, p0.5). The majority of these findings agree with the results obtained from comparing intakes of the abovementioned nutrients among cases and controls, indicating that there is a positive association between dietary diversity and nutrient intake.
As a matter of fact, adequacy of nutrient intake has been positively associated with the number of different foods consumed (Ferguson et al., 1993; Onyango et al., 1998; Bernstein et al., 2002). Results showed that even though the energy intakes were very low compared to recommended intakes (FAO/WHO/UNU, 2001), average nutrient intakes were above the recommended values for all nutrients except βcarotene and α-tocopherol (knowing that participants at the 90th percentile met the recommendations for β-carotene) suggesting a diet of high quality. As mentioned before, the study area is rural and fairly traditional. The diet is composed of traditional Lebanese dishes and personal field observations indicated very rare use of canned or processed products.
On the other hand, no significant difference was observed between the dietary diversity scores (DDS) of cases and controls neither when leafy greens (MannWhitney U test, p0.6) nor when wild leafy greens (Mann-Whitney U test, p0.06) were considered part of the food groups into which the food items were divided. This is probably due to the fact that all groups contained at least one staple food and that the DDS covered a period of three months. Food groups could not be based on nutrient content as most food items (apart from staples) contained two or more of the nutrients studied. The reason why the dietary diversity scores (DDS) were not significantly different between cases and controls but the food variety scores (FVS) were, could also be explained by a synergistic interaction between lutein and