عادات غذایی و بهزیستن ذهنی؛ یک گونه شناسی از دانشجویان در دانشگاه دولت شیلی

Abstract The purpose of this study was to distinguish and characterize university student typologies according to their life satisfaction and satisfaction with their food-related life. An online survey was applied between June and August 2013 in five state universities in Chile, to 369 university students (mean age = 20.9 years, SD = 2.27). The survey included the Health-related Quality of Life Index-4 (HRQOL), Satisfaction with Life Scale (SWLS), Satisfaction with Food-related Life Scale (SWFL), as well as questions about the place of residence, importance of food for well-being, frequency of meals in the place of residence and the frequency of consumption of eight food groups. A cluster analysis was used to determine student typologies. Three typologies of students were distinguished with significant differences in the average scores of the SWLS and SWFL scales, self-perception of health, days with mental health problems, number of days of health-related incapacity, place of residence, socioeconomic status, importance of food for well-being, frequency of breakfast and dinner in the place of residence, frequency of consumption of meat, milk, fruits and vegetables. It was found that most students with higher levels of life satisfaction and satisfaction with food-related life live with their parents, eat at home more frequently, report fewer health problems, have healthful eating habits and consider food very important for their well-being. Although it is necessary to promote or improve the campaigns that foster healthful eating in the entire university population, these campaigns must be specifically targeted to students who do not receive direct support from their families

Introduction While adult life satisfaction has been studied extensively, life satisfaction in younger people (late teens and early 20s) has only received attention more recently (Proctor, Linley, & Maltby, 2009). Research on satisfaction with food-related life has also emerged of late (Grunert, Dean, Raats, Nielsen, & Lumbers, 2007) and has been mainly circumscribed to adult samples, while knowledge regarding satisfaction with food-related life in youths and the variables that affect them is still scarce. Recent studies support that food is one of the important domains of life that affect an individual's life satisfaction (Grunert et al, 2007, Schnettler et al, 2013, Schnettler et al, 2013 and Schnettler et al, 2013), suggesting that satisfaction with food-related life is positively related to overall life satisfaction. This relation is associated to healthful eating behaviours and family interaction around food. Researchers have concluded that adults who eat more frequently with their families eat more healthfully and are more satisfied with their life and their food-related-life (Schnettler, Peña et al., 2013). However, these variables have not been studied in younger populations. Although there are several studies linking family meals and healthful eating habits in children and adolescents (Franko et al, 2013, Larson et al, 2007, Neumark-Sztainer et al, 2003, Nicklas et al, 2000 and Videon, Manning, 2003), the relationships between eating and well-being in young adulthood are less explored (Arnett, 2000). Some research results associate eating disorders with low levels of life satisfaction in young people (Halvorsen & Heyerdahl, 2006). Grant, Wardle, and Steptoe (2009) found a positive relation between a healthful diet and greater life satisfaction in university students in various countries. An exploratory study conducted with university students in southern Chile suggested that family support is associated with better eating habits, better health and greater satisfaction with life and food-related life (Schnettler, Denegri et al., 2013). Given that some researchers have concluded that life satisfaction is age-sensitive (Clench-Aas et al, 2011 and Gerstorf et al, 2008) it can also be expected that satisfaction with food-related life will change over the years. Regarding life satisfaction, while adults are increasingly concerned about avoiding negative experiences and focusing on health-related aspects (Clench-Aas et al., 2011), younger people focus on pleasant experiences (McMahan & Estes, 2012). For example, young people mistakenly believe that life satisfaction decreases with age, which can lead to hedonist and unhealthful behaviours, such as an unhealthful diet, in order to make the most out of life during youth (Garry & Lohan, 2011). Boelsma, Brink, Stafleu, and Hendricks (2010) and Ares, De Saldamando, Giménez, and Deliza (2014) have linked food and well-being in adults, which in turn associates to physical health, body functioning, intellectual capacity, positive emotions and social contact and relationships. Therefore, studying the relationship between life satisfaction, satisfaction with food-related life and eating habits in young people can provide important input to make a positive impact in their adult life. However, to our knowledge, the psychometric properties and the relation between the Satisfaction with Food-related Life scale (SWFL) (Grunert et al., 2007) and the Satisfaction with Life Scale (SWLS) (Diener, Emmons, Larsen, & Griffin, 1985) have not been assessed previously with university students. The study of food as a life domain in university students is particularly relevant because the period of university studies usually entails a change of lifestyle and increasing stress (Antúnez & Vinet, 2013), and the students become responsible for their meals (Cluskey, Grobe, 2009 and Verger et al, 2009). This critical stage in the development of eating habits will affect their future health and the health of their future families (Brown, Dresen, & Eggett, 2005). University students who live away from home engage in riskier eating behaviours because of the pressures of independence and hurried lifestyles (Barker, Galambos, 2007 and Brown et al, 2005), resulting in poor nutrition (Li et al., 2012). Furthermore, an inadequate diet during the university years could result in unfavourable physiological consequences that may lead to diet-related chronic diseases (Winkleby & Cubbin, 2004), and have negative psychological and social repercussions (Hidalgo, Hidalgo, Rasmussen, & Montaño, 2011). Various authors have described the nutritional vulnerability of university students whose eating habits are characterized by excessive consumption of high fat foods, sugars and salt, fast food and soft drinks, skipping meals frequently, taking a short time to eat, snacking, and consuming insufficient amounts of fruit, vegetables and fibre (Brannan et al, 2013 and Guthrie et al, 2002), all of which are linked to unhealthful diets (Guthrie et al., 2002). While the incidence of unhealthful eating behaviours among university students is well established, we know less about how such behaviours affect students' satisfaction with their food-related life and their overall life satisfaction. Some students may engage in unhealthful eating behaviours because of external pressures or lack of knowledge, but may be dissatisfied with their own behaviour, with negative consequences for their overall life satisfaction. Other students may enjoy their unhealthful behaviours because of sensory gratification or they feel that these behaviours fit their lifestyle. Insights into these relationships have important consequences for attempts to induce students to healthier eating habits, as students who are satisfied with their food-related life will be more difficult to address than students who are not. In this study, we focus on Chilean public university students' satisfaction with life and its relation to the food domain. Studying the relationship between satisfaction with life overall and in domains of importance to young people is relevant because high life satisfaction can help absorb the negative consequences of stress, psychological problems and various disorders (Proctor et al., 2009). In addition, since eating habits during the university years are linked to high prevalence of overweight and obesity (Aguilar-Ye et al, 2010 and Hidalgo et al, 2011), diet-related chronic diseases (Winkleby & Cubbin, 2004), and negative psychological and social repercussions (Hidalgo et al., 2011), identifying variables associated to students' satisfaction with food-related life may contribute to the development of strategies to prevent these negative effects. Therefore, the aim of the present study was to distinguish and characterize university student types according to their life satisfaction, satisfaction with their food-related life, sociodemographic characteristics, general health and eating habits. Also, the psychometric properties and the relation between the SWFL and the SWLS were evaluated by use of confirmatory factor analysis (CFA). This methodology tests whether a particular factor model is consistent with the data (Lévy, 2006). In this study we address four hypotheses: H1. Life satisfaction is positively related to the students' satisfaction with food-related life. H2. Life satisfaction and satisfaction with food-related life are positively related to the students' healthful eating habits. H3. Life satisfaction and satisfaction with food-related life are positively related to the frequency of family meal times. H4. Life satisfaction and satisfaction with food-related life are positively related to the students' better general health.

Results The mean age of the sample was 20.9 years (SD = 2.27); 53.7% were women; 95.4% resided in an urban area, 85.6% were of Chilean origin (Table 1). The sample obtained presents a similar composition to the population of university students enrolled throughout the country in 2013, in terms of gender, area of residence (CNED, 2014) and age (Navarrete, Candia, & Puchi, 2013). However, this sample shows a greater proportion of students with native background (Mapuche and Aymara) than the one reported by Blanco and Meneses (2011), corresponding to 6.7% of the general student population. This is due to the inclusion of the Tarapacá and de La Frontera universities in this study. Both institutions are located in the Arica and Parinacota region and La Araucanía region, respectively, which have the largest indigenous population in the country (Instituto Nacional de Estadísticas, 2005). Table 1. Socio-demographic characteristics of university students sample from Chile, August 2013 (n = 369). Characteristic Total Sex, %  Female 53.7  Male 46.3 Age, years Mean (SD) 20.9 (2.27) Zone of residence, %  Urban 95.4  Rural 4.6 Ethnic origin, %  Chilean 85.6  Ethnic origin (Mapuche, Aymara) 9.5  Others 4.9 Place of residence during study period, %  With parents the entire year 57.2  With parents the entire year although he/she travels for the day to attend class 10.6  With their parents only on weekends or for vacations 23.0  Independent of parents 9.2 Education level of the head of the household, %  Primary 13.8  Secondary 35.8  Tertiary 43.1  Postgraduate 7.3 Socioeconomic level, %  High and upper-middlea 21.4  Middle-middleb 28.2  Lower-middlec 24.1  Lowd 20.1  Very lowe 6.2 Note: The national currency values (Chilean pesos) were converted to US dollars using the average 2013 value (Ch$495.31/US$1). a High and upper-middle represents 7.2% of the population. The household head's education averages 16.2 years, which typically means he/she completed university studies. Monthly income in high and upper-middle homes ranges between US$3500 and $7200 or more. b Middle-middle represents 15.4% of the Chilean population. The household head's education averages 14 years, which typically means he/she completed technical studies or did not complete university studies. Monthly income in middle-middle homes ranges between US$1400 and $2500. c Lower-middle represents 22.4% of the population. The household head's education averages 11.6 years, which typically means he/she completed high school studies. Monthly income in lower-middle homes ranges between US$830 and $1050. d Low represents 34.8% of the population. The household head's education averages 7.7 years, which typically means he/she did not complete high school studies. Monthly income in low homes ranges between US$415 and $620. e Very low represents 20.3% of the population. The household head's education averages 3.7 years, which typically means he/she did not complete elementary school studies. Monthly income in very low homes is ≤ US$330. Table options The sample comprised mainly students living with their parents all year (57.2%) or living with their parents on weekends or for vacations (23.0%) (Table 1). The high proportion of students living with their parents during the year can be explained by the existence of universities in most regions of Chile, to cover the long geography of the country (Chile extends in the southern hemisphere between parallels 17°29′57″ S and 56°32′ S, in a 4270 km span). Since there is a public university plus several private universities in almost all regions of Chile, usually in the regional capital city, most students come from the same region where the institution is located. This is the case of four of the five universities considered in this study (Universidad de Tarapacá, Universidad de Talca, Universidad de La Frontera and Universidad de Magallanes). The largest proportion of students belonged to families where the head of the household had an educational level of high school (35.8) and university (43.1%). 28.2% of the sample belonged to the middle-middle SES and 24.1% to lower-middle (Table 1). In the first question from the HRQOL-4 (Hennessy et al., 1994), most students perceived their health as good (37.1%) or very good (32.5%). Students reported a high number of days with mental health problems (Mean = 7.5, SD = 8.4). 31.2% considered eating as very important for their well-being (31.2%), and 39.6% considerably important (Table 2). Table 2. Self-perception of health, days on which the student was affected by health problems and importance of eating for their well-being of university students sample from Chile, August 2013 (n = 369). Characteristic Total Self-perception of health (HRQOL-4), (%)  Very poor health 2.2  Fair health 22.5  Good health 37.1  Very good health 32.5  Excellent health 5.7 The number of days on which the respondent's physical health was not good in the last month (HRQOL-4), n Mean (SD) 4.7 (5.9) The number of days on which the respondent's mental health was not good in the last month (HRQOL-4), n Mean (SD) 7.5 (8.4) The number of days on which the person could not perform their usual activities due to health concerns (HRQOL-4), n Mean (SD) 2.8 (4.1) Importance of food for well-being, %  Not at all important 0.3  Hardly important 3.0  Slightly important 10.6  Very important 31.2  Considerably important 39.6  Completely important 15.4 Table options 53.1% of the students had breakfast daily in their residence, and they mainly had lunch daily (30.9%), or two or three times per week (35.5%) in that place. The highest proportion did not have dinner (35.9%) in their residence, whereas 25.8% had dinner there daily (Fig. 2). Most of the students ate bread daily (74.0%), cereals and pasta two or three times per week (57.5%), meat two or three times per week (50.1%), fish and seafood occasionally (50.7%), milk and dairy products daily (44.4%), fruit daily (29.3%) or two or three times per week (36.3%), vegetables daily (47.2%) and soft drinks daily (55.0%) (Fig. 3). Frequency of meals at the place of residence reported in a university students ... Fig. 2. Frequency of meals at the place of residence reported in a university students sample from Chile, August 2013 (n = 369). Figure options Full-size image (62 K) Fig. 3. Frequency of consumption of eight groups of foodstuffs and drinks of university students sample from Chile, August 2013 (n = 369). Figure options Following the CFA results, cluster analysis was performed, taking into account the sum score of only the remaining three items from each scale. Therefore, the theoretical score for each scale ranged between 3 and 18. The mean score of all participants (n = 369) was 12.6 (SD = 3.3) on the SWLS and 10.6 (SD = 3.3) on the SWFL. Cluster analysis resulted in the identification of three types of students. According to analysis of variance, the types differed significantly in the average scores of the SWLS and SWFL, the number of days with mental problems and the number of days on which the students could not perform their usual activities due to their health in the last month (P ≤ 0.001) (Table 3). According to the Crosstab procedure and Pearson's Chi2 test, the types also differed in the residence during the semester, SES, importance assigned to eating for personal well-being (P ≤ 0.05), and in the self-perception of health (P ≤ 0.001) (Table 4). They differed in the frequency with which the students had breakfast (P = 0.000) and dinner (P = 0.013) in their residence (Fig. 4), the frequency of consumption of meat (P = 0.002), milk and dairy products (P = 0.005), fruits (P = 0.004) and vegetables (P = 0.015) (Fig. 5). No significant differences were observed between typologies according to age, gender, area of residence, education level of the head of the household, ethnic origin, year of enrolment, or the university attended; nor regarding the number of days with physical health problems in the last month, frequency with which the students eat lunch at their residency, frequency of consumption of bread, cereals and pasta, fish and sea food, and soft drinks (P > 0.1). Table 3. Satisfaction with life (SWLS), Satisfaction with food-related life (SWFL) mean scores and days on which the student was affected by mental health problems and could not perform their usual activities due to health concerns of the groups obtained using cluster analysis in university students from Chile, August 2013 (n = 369). Group 1 (n = 132) Group 2 (n = 142) Group 3 (n = 95) F P-value SWLS Mean (SD) 15.01 a (1.65) 13.26 b (1.99) 7.30 c (1.66) 301.986 0.000** SWFL Mean (SD) 13.96 a (1.62) 8.80 b (1.72) 7.19 b (2.51) 283.445 0.000** The number of days on which the respondent's mental health was not good in the last month, n Mean (SD) 4.79 b (5.63) 6.99 b (7.64) 13.49 a (8.33) 18.746 0.000** The number of days on which the person could not perform their usual activities due to their health, n Mean (SD) 2.27 b (3.54) 2.31 b (3.28) 4.58 a (3.32) 5.876 0.001** Note: Theoretical scores of SWLS and SWFL scales ranged between 3 and 18. Group 1 “Satisfied with their life and their food-related life”. Group 2 “Moderately satisfied with their life, unsatisfied with their food-related life”. Group 3 “Unsatisfied with their life and their food-related life”. ** Significant at P ≤ .001. Different letters (a, b, c) in rows indicate statically significant differences according to Dunnett's T3 Comparison test for non homogenous variances. Table options Table 4. Characteristics (%) with significant differences of the groups obtained using cluster analysis in university students from Chile, August 2013 (n = 369). Group 1 (n = 132) Group 2 (n = 142) Group 3 (n = 95) Place of residence during study period P = .010  With parents the entire year 68.9 50.7 50.7  With parents the entire year but commutes for the day to attend class 7.7 12.6 7.5  With parents on weekends and on vacation 18.9 25.4 18.9  Does not live with parents 4.5 11.3 22.9 Self-perception of health P = .000  Very poor health 0.1 2.1 7.5  Fair health 12.8 20.4 46.3  Good health 28.0 45.1 32.8  Very good health 50.8 26.8 11.9  Excellent health 8.3 5.6 1.5 Socioeconomic status P = .014  High and upper middle 25.8 21.1 11.9  Middle-middle 27.3 34.5 19.4  Lower middle 25.0 23.2 25.4  Low 19.7 14.1 34.3  Very low 2.3 7.0 9.0 Importance of food for well-being P = .031  Not at all important 0.8 5.1 0.1  Hardly important 2.3 16.0 5.9  Slightly important 3.8 13.4 13.4  Very important 25.8 30.1 40.3  Considerably important 50.0 25.0 25.4  Completely important 17.4 13.4 14.9 P values obtained by Pearson χ2 test. Group 1 “Satisfied with their life and their food-related life”. Group 2 “Moderately satisfied with their life, unsatisfied with their food-related life”. Group 3 “Unsatisfied with their life and their food-related life”. Table options Frequency of meals at the place of residence with significant differences of the ... Fig. 4. Frequency of meals at the place of residence with significant differences of the groups obtained using cluster analysis in university students from Chile, August 2013 (n = 369). Group 1 “Satisfied with their life and their food-related life”. Group 2 “Moderately satisfied with their life, unsatisfied with their food-related life”. Group 3 “Unsatisfied with their life and their food-related life”. Figure options Frequency of consumption of foodstuffs with significant differences of the ... Fig. 5. Frequency of consumption of foodstuffs with significant differences of the groups obtained using cluster analysis in university students from Chile, August 2013 (n = 369). Group 1 “Satisfied with their life and their food-related life”. Group 2 “Moderately satisfied with their life, unsatisfied with their food-related life”. Group 3 “Unsatisfied with their life and their food-related life”. Figure options Group 1 “satisfied with their life and their food-related life” (35.8% of the sample, n = 132) had significantly higher scores than the other groups on the SWLS and SWFL scales. This group registered the lowest number of days affected by mental health problems and limitations on daily activities due to health problems, although it did not differ statistically from Group 2 (Table 3). Based on the adjusted standardized residuals analysis, Group 1 had a greater presence of students who lived with their parents all year (68.9%), perceived their health was very good (50.8%), considered eating was considerably important for their well-being (50.0%) (Table 4), tended to have breakfast (66.7%) and dinner (37.9%) daily in the place where they lived (Fig. 4), and consumed milk and dairy products (54.5%), fruits (41.7%) and vegetables (57.6%) daily (Fig. 5). Group 2 “moderately satisfied with their life, unsatisfied with their food-related life” (38.4%, n = 142) had a significantly higher score than Group 3 on the SWLS, but did not differ statistically from Group 3 in the score on the SWFL (Table 3). Based on the adjusted standardized residuals analysis, this group had greater presence of students with a good self-perception of their health (45.1%), were middle-middle on the SES (34.5%), and considered eating as not at all important (5.1%) or hardly important (16.0%) for their well-being (Table 4). Group 2 had the highest proportion of students who had breakfast where they lived only two to three times per week (22.5%) (Fig. 4), and a lower presence of students who ate fruit daily (23.2%) (Fig. 5). Group 3 “unsatisfied with their life and their food-related life” (25.7%, n = 95) had a significantly lower score on the SWLS compared to the other groups. Members of this group reported significantly more days affected by mental health problems and more days when they could not perform their normal activities due to health reasons (Table 3). Based on the adjusted standardized residuals analysis, Group 3 contained a higher proportion of students who lived away from the family home (22.9%), had a self-perception of their health being very poor (7.5%) and fair (46.3%), and belonged to the low SES (34.3%) (Table 4). In this typology, a greater proportion of students skipped breakfast (20.9%) and dinner (48.5%) (Fig. 4), and only occasionally ate meat (14.9%), milk and dairy products (20.9%), fruits (28.4%) and vegetables (14.9%) (Fig. 5).