Actitudes de adolescentes hacia la salud : evaluación de un programa escolar de promoción de la salud en Sevilla , España Attitudes of adolescents regarding health : evaluation of a school-based health promotion program in Seville , Spain

The study examined differences in attitudes regarding health within students in their fourth year of mandatory secondary school in Seville, Spain. The students were divided into two groups according to their participation in the “Forma Joven” health promotion program. A descriptive and bivariate analysis was conducted. As there were no significant differences in the socio-demographic characteristics of the two groups, the groups were considered homogeneous. Attitudes were also similar in both groups. Statistically significant differences were only found in the level of danger assigned to two of the thirteen transit situations explored: “not using a seatbelt” and “entering a vehicle when the driver has been drinking,” with the lower level of risk corresponding to the group of students participating in the program. These findings are consistent with the results of previous studies and invites reflection upon the effectiveness of such programs. In the case of the “Forma Joven” program, the ineffectiveness could be attributable to a lack of systematization in the program’s implementation. In the future, interventions to increase the effectiveness of the program should be proposed. kEy wordS Health Promotion; Adolescents; Program Evaluation; Risk Behavior; Attitude to Health. Actitudes de adolescentes hacia la salud: evaluación de un programa escolar de promoción de la salud en Sevilla, España Attitudes of adolescents regarding health: evaluation of a school-based health promotion program in Seville, Spain Lima Serrano, Marta1; Neves de Jesus, Saul2; Lima Rodríguez, Joaquín Salvador3 1Licenciada en Antropología Social y Cultural. Diploma de Estudios Avanzados en Enfermería. Profesora Ayudante de Enfermería, Universidad de Sevilla, España. mlima@us.es 2Licenciado en Psicología. Doctor en Psicología de la Educación. Profesor Catedrático de Psicología, Universidad de Algarve, Portugal. snjesus@ualg.pt 3Licenciado en Antropología Social y Cultural. Doctor en Enfermería. Profesor Titular de Enfermería, Universidad de Sevilla, España. joaquinlima@us.es


INTRODUCTION
Traffic accidents were the leading cause of death in the age group of 15 to 24-year-olds in Andalusia (Spain) in 2001 (1).In terms of the number of victims involved, for each fatality approximately 6 other people were severely wounded and 29 suffered less serious injuries.A higher incidence of accidents is associated with the consumption of alcohol and other drugs (2).The consumption of addictive substances also has a high cost of disease in social and economic terms (3)(4)(5).In Andalusia, a research study conducted in 2006 involving 11 to 17-year-old adolescents revealed that the average age of initiation of tobacco and alcohol consumption was 13.5 years, and that 36.2% of the population studied had consumed tobacco and 57.6% alcohol.Consumption of illegal substances was lower: 8.5% had at one time consumed cannabis and about 1% had consumed other drugs (6).
In terms of sexual behaviors, 20% of Andalusian adolescents between 11 and 17 years of age did not use condoms during sex whereas 5.2% did not use any contraceptive method.Two and a half percent of adolescent girls had had at least one pregnancy, with 15-16 years being the age of highest risk (7).A growing trend has been observed in the voluntary termination of pregnancy (8).AIdS was the third leading cause of death in 2001 for young people aged 25 to 34 years, although infection had occurred several years before death (1).
obesity is considered the epidemic of the 21st century (9).In 2003, the prevalence of overweight between people aged 16 to 24 was 15.4% and that of obesity was 3.1% (10).
All of these situations are linked to relational and behavioral aspects and their reduction requires strategies for health promotion and risk prevention involving different sectors (1).
most health promotion programs at schools are focused on a specific problem (drug consumption, pregnancy prevention, sexually transmitted diseases, etc.), frequently as a product of uncoordinated, fragmented and inefficient initiatives.Programs should be implemented with a more comprehensive focus, centering on the positive development of children and adolescents in order to generate cognitive, personal, emotional, moral and social competencies that allow for the "empowerment" of these population groups and their positive contribution to the community (11)(12)(13).
during the 2001-2002 academic year, the program "Forma Joven" was started in order to promote the health of youth and prevent risk behaviors.Based in an intersectoral approach, the main goal of the program was to establish points of service in places frequented by adolescents, in particular high schools, with the target population being adolescents enrolled in their first to fourth year of compulsory secondary education (3).during the 2009-2010 academic year, there were 778 "Forma Joven" program locations in Andalusia, 41 in the capital city of Seville and 113 in the province of Seville (14,15).In these locations a series of health promotion and education activities were developed by a team of health and education professionals, usually made up of a nurse from a primary care center and a counselor from the institution of secondary education, joined by other social agents and youth acting as mediators.
The main areas of intervention of the program are: lifestyle (balanced diet and physical activity, addictive consumption, road safety), sexuality and relationships, mental health, peer interaction and violence prevention (1).The team proposes primarily extracurricular activities, the central activity being individual or group counseling for youth who, by appointment and anonymously, pose their questions and receive information and advice.If a special problem is identified, follow-up is carried out and if need be the student may be referred to other institutions.Counseling is held on a weekly, biweekly or monthly basis, depending on the school and the volume of the demand.The rest of the activities primarily consist of workshops with defined groups (for example, a group of students in their fourth year of compulsory secondary education), addressing topics related to the main areas of intervention, in accordance the needs identified.An expository and participatory methodology is used, and, sometimes, the issue is previously addressed by peer tutors.
Health campaigns and contests, among other activities, are also organized, with the frequent participation of social actors such as youth or women's delegations from the local government, nongovernmental organizations in the area, centers for drug addiction, the police force, the civil guard, etc.These activities are not usually evaluated; if they are, it is by way of the assessments or satisfaction surveys of the peer tutors, who are asked about the experience, their suggestions for improvement, etc.Although there is a record of data collection which would allow for a monitoring and evaluation process, no evaluations have been carried regarding the impact of the program "Forma Joven" on the health of participants (1).An evaluation of the effect or impact of a program allows for the examination of the improvements achieved within the population as a result of the intervention (16).In the case of the "Forma Joven" program, this would mean youth developing health-promoting attitudes and behaviors towards the major risks they face.
The objective of this work was to ascertain the attitudes towards food, sexuality, addictive substances and road safety of adolescents in their fourth year of compulsory secondary education in the urban area of Seville, Spain, and to determine whether there are differences in the attitudes of students who were part of the intervention group -participants in the program "Forma Joven" -and those of students who were part of the control group and therefore did not participate in the program.

MATERIALS AND METHODS
An evaluative, quasi-experimental and postintervention design was used to analyze the effects of the "Forma Joven" program, including an intervention group that had been part of the program and a control group that had not been part of it (17).With the program under evaluation already running, its impact could be assessed by comparing the participating students with a control group, attempting to control statistically for certain variables that could influence the acquisition of risk behaviors (18).measures were taken to perform an "internal control" or group pairing so that the groups were as similar as possible to each other (19).
Considering the way in which the program is carried out, it was assumed that the student group receiving the greatest effect was made up of those in their fourth year of compulsory secondary education.The study population was composed of 15 to 17-year-old adolescents enrolled in secondary schools in the capital city of Seville.The sample was taken from the 50 state-run institutions (20), 38 of which participated in the program (15).Eight institutions were selected, four from the intervention group and four from the control group.The following inclusion criteria were established: Heterogeneity: the presence of the "Forma Joven" program in the schools included in the intervention group, or the absence of the program in the control group.Homogeneity: state-run high-schools; fourthyear students in compulsory secondary education; the institutions of both groups should belong to the same geographical area, considering neighborhood distribution.
In the case of the schools in the intervention group, the conditions established were that the program had to have been implemented at least two academic years prior to the start of the research study and that there had to be records monitoring the process.The institutes located outside the urban area and those whose authorities refused to participate were excluded.The sample was obtained randomly using cluster sampling, choosing a fourth-year group from each school.The initial sample included 166 adolescents, from whom the information was collected.After the exclusion of those whose participation was inadequate, due to inconsistency in their responses or incompleteness of responses (less than 75%), the final distribution included 73 adolescents in the intervention group and 73 in the control group (N=146).
The independent variable was the presence or absence of the program in the schools.Attitudes toward food, sexuality, addictive substances and road safety were the dependent variables explored.
According to the risk and protective factors for the acquisition of risk behaviors that were identified in the literature, it was decided to control for the following variables: age, sex/gender, educational level of parents, perceived socioeconomic status and family dynamics (21)(22)(23)(24)(25).
As no validated instrument consistent with the objective of the study was found, a questionnaire of our own elaboration was used, which was created by taking into consideration both the theoretical framework and previously validated questionnaires (26)(27)(28).The survey was pilot tested on 22 adolescents with similar characteristics to the participants in order to analyze the appropriateness, understandability, length, and possible existence of fatigue.The final instrument included 85 items, divided into five subscales.The first three assessed dichotomically (agreement/disagreement) attitudes toward food (such as "fast food allows for better use of time"), sexuality (such as "I prefer sexual intercourse involving penetration, even though I could get a disease") and addictive substances ("people who drink are more interesting").
Based on the consideration that the perception of risk, defined as the degree to which a behavior is attributed to an alleged health hazard, is involved in the adoption of behaviors (29,30), two four-point Likert scales were constructed in order to study the degree of risk assigned to the consumption of addictive substances as well as the degree of risk assigned to specific road safety situations (for example, "entering a vehicle when the driver has been drinking").
To control group homogeneity, five questions were included on socio-demographic characteristics and the Apgar family test for the study of family function was used (31).
The study was approved by the universidad de Sevilla Experimentation Committee.Anonymity, information confidentiality and data protection were guaranteed by requesting informed consent from participants, who were considered competent to provide such consent given the type of study and their age.
data collection was conducted between may and June 2009, following the indications of the study Health Behaviour in School-Aged Children, which state that: students should complete to the questionnaire themselves, anonymity should be guaranteed, and questionnaires should be conducted within the school setting by duly trained personnel.Therefore, the questionnaire was provided to students by a researcher who explained to them the nature of the study and how to complete the questionnaire.Students then completed the surveys, which were self-administered.
A descriptive analysis of the research study variables and a bivariate analysis to compare the groups were performed.The Pearson Chi 2 test was used (χ 2 ) to assess the statistical significance of differences in proportion (p<0.05), to check the homogeneity of the groups in relation to the controlled variables and to identify differences in terms of the independent variable (presence/absence of the program in the high school).The effect size was measured through the contingency coefficient (rφ): low, rφ=0.1;medium, rφ=0.3;high rφ=0.5.missing data was not included in the analysis.The statistical program SPSS 17.0 was used.

RESULTS
When testing group homogeneity, there were no statistically significant differences in the distribution of sociodemographic variables or in the family function of the participants; thus, they were considered homogeneous (Table 1).In order to compare the variables "level of education attained by the father" and "level of education attained by the mother," the "none" category was eliminated (the expected frequency was less than 5%), and within the variable "perception of the family economy" the categories "we have economic problems at home" and "we manage, but we barely make it to the end of the month" were grouped together.
There were no statistically significant differences between the intervention group and the control group in any of the analyzed attitudes regarding food, sexuality, and addictive substances.
In both groups, the majority of the attitudes analyzed were positive.However, with regard to food consumption, the fact that more than half of the students felt that "fast food allows for better use of time" was highlighted as negative.Furthermore, there was an important percentage of students those who thought it fine to eat sweets or baked goods daily, as well as who stated that "when someone is on a diet it is normal to skip a meal" (Table 2).
Regarding sexuality no statistically significant differences were found; however, a trend in some of the items was found.Surprisingly, the percentage of agreement in the students belonging to the intervention group was higher regarding the following statements: "my boyfriend/girlfriend does not like to use condoms," "I prefer sexual intercourse without a condom, even if I risk getting a disease," "I prefer sexual intercourse involving penetration, even though I could get a disease." With respect to attitudes regarding sexuality, in both groups, the fact that an important number of students thought that "using condoms is uncomfortable because they reduce sensitivity" or that "men are responsible for protection" was highlighted as negative.
Regarding the consumption of addictive substances, the significant number of adolescents who felt that "smoking is trendy" was highlighted as negative whereas the fact that the tobacco consumption of others bothered most of the adolescents was highlighted as positive.In the case of alcohol intake, the number of students having favorable -and therefore negative -attitudes toward consumption was greater. of these, the following attitudes stand out: "alcohol makes you feel good," "it helps you make friends," and "it is trendy."only a small percentage was bothered by the alcohol consumption of others (Table 4).
Regarding the use of tranquilizers, the percentages of favorable attitudes toward consumption were lower.The percentage of students who felt that "they make you feel good" was highlighted as negative, as was the fact that only 28.8% of the adolescents were bothered by their consumption.Regarding hashish consumption, the percentages of students with attitudes such as "it makes you feel good" and "it is trendy" were highlighted as negative as well as the low percentage of students bothered by its consumption (Table 5).
Regarding cocaine, 9.6% of the intervention group and 17.8% of the control group (p=0.114)stated that "it was trendy" and 41.1% of the intervention group and 37.7% of the control group were bothered by its consumption.Regarding other substances, 1.4% of the intervention group and 8.2% of the control group (p=0.058)thought that consuming hallucinogens was trendy and only 35.6% of the intervention group and 38.6% of the control group were bothered by their use, while 43.8% of the intervention group and 35.6% of the control group were bothered by ecstasy use.Substances mostly considered "not at all" or "a bit" dangerous were tranquilizers, tobacco, alcohol and hashish.other illegal drugs (such as cocaine, ecstasy and hallucinogens) were considered "fairly" or "very" dangerous by most participants (Table 6).
The road situations considered as "not at all" or "a bit" dangerous were the following: not wearing a helmet when biking, crossing the street where it is prohibited to do so, driving on a badly maintained road, and traveling with a driver who is talking on a cell phone (Table 7).Statistically, the only significant difference noted between the two groups was the degree of risk assigned to two of the thirteen road situations explored: not wearing a seat belt (p=0.002;rφ=0.25), and entering a vehicle when the driver has been drinking (p=0.028;

rφ=0.19
).There was also a trend observed in the item "not wearing a helmet when riding a motorcycle" (p=0.060).The degree of danger assigned was lower in the intervention group.

DISCUSSION
The results show that participants, in general, had favorable attitudes towards the adoption of healthy behaviors regarding food, sexuality, substance abuse and road safety.However, in relation to food, a large percentage of students showed positive opinions towards the daily consumption of candy or pastries and skipping meals while dieting.Previous studies suggest an increasing trend in the consumption of sweets in the Spanish school-aged population (32,33).
Regarding sexuality, unfavorable attitudes toward condom use were observed, intercourse was highly valued as an expression of sexuality, and sexist attitudes were also observed.This could indicate the existence of a masculine pattern in the participants' attitude toward sexuality (34).most addictive substances were considered "trendy."Alcohol showed the highest percentage of agreement with attitudes of acceptance.A similar research study conducted in Barcelona (35) also demonstrated a high percentage of attitudes of acceptance towards alcohol.In that study, 30.6% answered affirmatively when asked if alcohol made them feel happier, 52.2% felt that alcohol made parties more fun and 68.2% considered that it was alright to advertise alcoholic beverages.moreover, these attitudes relate to the findings of previous studies, which concluded that alcohol is the substance most consumed among adolescents from Andalusia and Spain, much more consumed other drugs.Thus, in 2008, 81.2% of secondary school students aged 14 to 18 years had consumed alcoholic beverages at some point in their lives (29).In Andalusia, in 2006, 57.6% of students between 11 and 17 years of age had consumed alcoholic beverages (8).
despite the fact that most of students surveyed were bothered by others smoking tobacco, the results of this item were not very encouraging regarding substances such as alcohol and other drugs.This may reflect a high degree of acceptance of the consumption of these substances among the study participants.
The study found that hashish and legal drugs were the substances considered least dangerous.on the other hand, illegal drugs were considered dangerous by the majority.This result is similar to the findings of a study by Nebot et al. (35), except in the case of hashish, which was largely considered very or moderately dangerous in that study (92.1%).The degree of risk assigned to addictive substances or the perception of risk regarding their consumption is indicated as a risk factor for their use (29,30), reflected in the fact that legal drugs and hashish are the substances most commonly consumed among youth in Andalusia and Spain.This situation could be similar to the road situations considered less dangerous by the study participants, such as not wearing a helmet when riding a bicycle or crossing the street where it is prohibited to do so.
In the homogeneity tests no significant differences in sociodemographic characteristics and family dynamics were found.Trends were observed that differentiated some of the studied attitudes by group, sometimes in favor of the intervention group and sometimes in favor of the control group, but statistically significant differences were only observed in the perceived risk of two of the thirteen road safety risk situations explored, which did not reach a medium effect size (rφ=0.3).These differences could perhaps be related to the existence of other confounding variables which are not considered in this study.In general, at schools, even if the program "Forma Joven" is not present, other types of interventions are carried out.In the case of traffic accident prevention, social institutions such as the Association for the Study of Spinal Cord Injury (Asociación para el Estudio de la Lesión Medular Espinal) or other institutions such as the Traffic Civil Guard often visit schools.The investigation included "naturally formed" populations and in some of the schools belonging to the control group these issues may have been addressed, helping to improve the students' perception of risk.It would be interesting in future studies to explore the degree of implementation of health promotion interventions in schools to improve the understanding of these results.
Programs for health promotion and prevention of risk behaviors often do not significantly influence attitudes (19,35,36).Although families and peer groups are considered primary agents in the socialization of healthy lifestyles (37) and working with these groups is part of the principles and actions of the "Forma Joven" program, during the planning interviews held within our study, we observed that in the schools implementing the program there was no participation of parents or mediators.
Fernandez et al. (38) conducted a systematic review to identify the key characteristics of effective prevention programs for adolescents, highlighting as important the elements and contents to be addressed, the methodology to be used (mainly active), the presence of peer mediators and families, the number and type of sessions, and a rigorous evaluation pre-and post-intervention to reflect the positive effects.However, these elements were not found in a systematic way in the schools within our study that carry out the "Forma Joven" program.
This opens a debate regarding the influence of these situations on the results of the study and the impact of the program.It could be argued that there is a gap between the principles and actions proposed by the program "Forma Joven" and the reality observed in the studied schools in terms of the activities carried out and the social agents involved, which could be related to a lack of systematization during the implementation process.
The study was conducted post-exposure, as the program had been underway for a few years; therefore, no prior information on students' attitudes was available.For this reason, a control group was included in the study, and some variables that could be influencing attitudes were controlled for statistically (17)(18)(19).Furthermore, the intervention and control groups were matched by geographical area, educational level and type of school in which they had been enrolled.In subsequent studies it would be advisable to use a larger sample size and supplement the study with qualitative methodology in order to inquire into the discourses of those involved in the program (professionals, families and adolescents).Such inquiries could help to understand the difficulties involved in program implementation, proposals for improvement and opportunities for promoting adolescent health (41), so as to generate interventions that promote the positive development of adolescents, contributing to overcoming the weaknesses of the program and increasing its effectiveness.
The study has met the bioethical principles of beneficence, nonmaleficence, autonomy and justice.The latter can be infringed in community trials, when applying an intervention or program in a discriminatory way to a subgroup of the population, despite its potential benefits for the whole population (17,40).Previously, the participating schools had voluntarily decided whether or not to participate in the "Forma Joven" program without interfering in its implementation.

Table 1 .
Sociodemographic characteristics of adolescents participating in the study, by intervention and control group.Percentages and absolute frequency.Seville, 2009.
Source: Own elaboration from data collected in the research study.a χ2 was calculated without including missing data.df = Degrees of freedom.universidad Nacional de Lanús | Salud Colectiva | English Edition ISSN 2250-5334 | E-ISSN 1851-8265| ISSN-L 1669-2381

Table 2 .
Degree of agreement with attitudes toward food consumption in the study participants, by intervention and control group.Percentages and absolute frequency.Seville, 2009.
Source: Own elaboration from data obtained in the research study.a χ2 was calculated without including missing data.*50% of the cells have an expected frequency of less than 5. Fisher's exact test was used.df = Degrees of freedom.

Table 3 .
Degree of agreement with attitudes toward sexuality in the study participants, by intervention and control group.Percentages and absolute frequency.Seville, 2009.50% of the cells have an expected frequency of less than 5. Fisher's exact test was used.df = Degrees of freedom.STD = sexually transmitted disease.

Table 4 .
Degree of agreement with attitudes toward tobacco and alcohol consumption, by intervention and control group.Percentages and absolute frequencies.Seville, 2009.
Source: Own elaboration from the data obtained in the research study.a χ2 was calculated without including missing data.*50% of the cells have an expected frequency of less than 5. Fisher's exact test was used.df = Degrees of freedom.

Table 5 .
Degree of agreement with attitudes towards tranquilizers and hashish in the study participants, by intervention and control group.Percentages and absolute frequencies.Seville, 2009.

73) Control group (n = 73) χ 2 (df, N) a p-value Intervention group (n = 73) Control group (n = 73) χ 2 (df, N) a p-value
Source: Own elaboration from data obtained in the research study.a χ2 was calculated without including missing data.*50% of the cells have an expected frequency of less than 5. Fisher's exact test was used.df = Degrees of freedom.

Table 6 .
Degree of dangerousness assigned to addictive substances by the study participants, by intervention and control group.Percentages and absolute frequencies.Seville, 2009.50% of the cells have an expected frequency of less than 5. Fisher's exact test was used.df = Degrees of freedom.

Table 7 .
Degree of danger assigned by the participants to situations related to road safety, by intervention and control group.Percentages and absolute frequencies.Seville, 2009.