Selective Prevention for Anxiety and Resilience Promotion : Outcomes of an Anxiety Prevention and Resilience Program with Girls at Risk

Research has shown that the number of children and adolescents experiencing emotional difficulties such as anxiety is escalating, especially for those groups at risk. Living in an orphanage has been considered a risk factor for delays in an individual ́s social, emotional and behavioral development. The aim of this study was to evaluate the effectiveness of the Spanish version of the FRIENDS program, a CBT-based resilience program, with 57 girls from low SES background that were living in an orphanage. Participants received the program for 10 consecutive weeks and preand post-test measures were collected. Measures evaluated participants ́ anxiety symptoms, level of self-concept, hope, coping skills, and psychosocial difficulties. Social validity was also assessed. Results showed positive changes, including a decrease in anxiety symptoms and psychosocial difficulties, and an increase in their proactive coping skills. Several subscales and items of the self-concept and hope outcome measures also reported statistically significant improvements. Participants and parents/caregivers reported that the program was both enjoyable and useful. Implications of the findings and directions for further research are discussed. Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) 2

simply absent, developing brain circuits can be disrupted, and insecure attachments could translate into future emotional and behavioral difficulties (Bowlby, 1973).
The FRIENDS program is a resilience program, developed by Dr. Paula Barrett, which uses cognitive-behavioral techniques for the prevention and early intervention of anxiety.There are three developmentally tailored versions of the program for children ages 4-7 (Barrett, 2011a;Barrett, 2011b), children ages 8-11 (Barrett, 2011c;Barrett, 2011d), and adolescents ages 12 to 16 (Barrett, 2011e;Barrett, 2011f).The program has been implemented in different countries and at all levels of prevention.For the selective level of prevention the following "at risk" groups have been studied: non-English speaking immigrants, African American children who have been exposed to community violence, children and youth from socioeconomically disadvantaged communities, children who reported high levels of anxiety sensitivity, students exhibiting anxious symptomatology, low mood, and low self-esteem, and children living in an orphanage (Barrett, Moore, & Sonderegger, 2000;Barrett, Sonderegger, & Sonderegger, 2001;Barrett, Sonderegger, & Xenos, 2003;Cooley-Quille, Boyd, & Grados, 2004;Cooley-Strickland, Griffin, Otte, & Ko, 2013;Gallegos, Rodriguez, Gomez, Rabelo, & Gutierrez, 2012;Liddle & Macmillan, 2010;Siu, 2007;Stopa, Barrett, & Golingi, 2010;Tortella-Felliu, Servera, Balle, & Fullana, 2004).In general, positive results have been reported after receiving the program such as: an increase in proactive coping skills, self-esteem, resilience, social skills, and positive future outlook.Results have also reported a decrease in anxiety and depressive symptoms, conflict with peers, and conduct problems.
There are some studies that have implemented programs with the aim of improving the physical conditions and the social and emotional development of children and youth living in orphanages (Ssewamala, Neilands, Waldfogel, & Ismayilova, 2012: McCall et al., 2010: Schnek, 2009;Groark, Muhamedrahimov, Palmov, Nikiforova, & McCall, 2005), but most of them were delivered to the caregivers not to children.There are two studies that that have implemented programs with the children.The study by Kumakech, Cantor-Graae, Maling,and Bajunirwe, (2009) evaluated the effect of a school-based peer-group support intervention combined with periodic somatic health assessments and treatment on the psychological well-being of AIDS orphans from Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) Uganda.Results showed that participants from the intervention group, in comparison with those from the control group, showed significant improvement in depression, anger, and anxiety.Gallegos et al. (2012) conducted a study to evaluate the effectiveness of the Spanish version of the FRIENDS for Life program with 10 girls, ages 9 to 10, living in a Mexican orphanage.Pre and post measures evaluated anxiety, depression, coping skills, self-concept, self-esteem and optimism.Results showed positive changes especially in their optimism and self-concept.Particular subscales and items also reported improvements such as a decrease in worry, physiological symptoms of anxiety, and negative mood, and an increase in self-esteem at home and with peers.
Participants´ evaluated the program as enjoyable and useful; and they rated "changing negative unhelpful thoughts to positive thoughts" and the "coping step plan" as particularly useful (Gallegos et al., 2012).
To extend current knowledge, the present study included a larger sample with a broader age range, and also included the evaluation of participants´ psychosocial difficulties and social validity as rated by caregivers.Two research questions guided this study: (1) What is the effect of the Spanish version of the FRIENDS program on girls´ anxiety, self-concept, hope, coping skills, and psychosocial difficulties?It was hypothesized that the self-concept, hope, proactive coping skills and pro social behavior would increase and that girls would report less anxiety symptoms and psychosocial difficulties after the intervention.( 2) To what extent girls and their caregivers were satisfied with the program and which of the skills they found most useful?It was hypothesized that girls and caregivers would enjoy the program and would find the skills learned to be useful to cope with daily life stressors.

Method
A one group pretest-posttest design was employed to address the research questions.
The independent variable was the Spanish version of the FRIENDS program and the dependent variables were: anxiety, self-concept, hope, coping skills, and psychosocial difficulties.Social validity was also evaluated.

Participants
Participants were 57 girls', ages 5 to 15 years old, that came from a low socio-economic background and were living in an orphanage in Mexico.The mean age of participants was 10.95 (SD= 2.94).Girls were living in the orphanage for several reasons: some of them were abandoned by their parents or have lost their parents, some of them came from dysfunctional families with situations such as drug abuse, physical and psychological abuse, and some of them have parents or close relatives that visit them on the weekends but do not have the resources to take care of them at home.
Participants in this study also included the caregivers, referring to the relative/parent who took care of them during the weekend, and for the ones that stayed all week in the orphanage, the caregiver was one of the nuns working in the orphanage.The city chosen for this study is one of the three cities in Mexico with the highest prevalence rate for anxiety disorders (Medina-Mora, et al., 2003).

Measures
Five measures were administered collectively to all participants to assess anxiety, selfconcept, hope, coping skills and psychosocial difficulties.Social validity was also evaluated.(Reynolds & Richmond, 1985).Escala de Ansiedad Manifiesta en Niños Revisada (CMAR-R) is the Spanish version of the Revised Children's Manifest Anxiety Scale, a 37-item self-report measure of trait anxiety in children and adolescents ages 6 to 19.A total of 28 items pertain to trait anxiety, with a further of nine items assessing social desirability.Items are presented as a series of statements, which participants are required to endorse as being either true or not true of them, by indicating either "yes" (1) or "no" (0) respectively.A total score is derived by summing together the responses for each item, with a higher score indicating greater levels of anxiety.The scale has three anxiety subscales: physiological anxiety, worry, and social anxiety.The CMAS-R has sound psychometric properties including high Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) internal consistency and moderate test-retest reliability (Reynolds & Richmond, 1985).
Piers-Harris Children´s Self-Concept Scale (Piers, 1984).Piers Harris 2: Lo Que Pienso de Mi Mismo is the Spanish version of the Piers-Harris Children´s Self-Concept Scale that was designed to examine the self-attitudes of children ages 8 -19.The selfreported measure assesses six aspects of a child´s self-concept: behavior, intellectual and school status, physical appearance and attributes, anxiety, popularity, and happiness and satisfaction.The instrument is a 60-item inventory consisting of short sentences for which the child answers yes or no.The items describe children´s feelings about themselves and about the reactions of others toward them; higher scores indicate a better self-concept.Each positive response is scored with 1 point and each negative response with 0 points.About half of the 60 statements indicate high self-concept and half are low self-concept.High scores indicate a better self-concept.CSCS total scale internal consistency ranges from 0.88 to 0.94 with stability ranging from 0.42 to 0.96.CSCS subscale internal consistency ranges from 0.73 to 0.81 (Bracken, Bunch, Keith, & Keith, 2000;Piers, 1984).Scores on the CSCS have shown adequate test-retest reliability (r = 0.80) and convergent validity (r = 0.61) with other self-concept instruments such as Multidimensional Self-Concept Scale (Piers, 1984).A reliability index of α=.938 was found (Cronbach´s Alfa) in this study.
The Children´s Hope Scale (Snyder et al., 1997) is a self-report measure designed to measure children´s dispositional hope.The measure was translated into Spanish for this study.The measure was developed for use with children 8-16 years and consists of six items, three of which assess agency thoughts and three of them assess pathways thoughts.In response to each item, the children are giving the 6option continuum: "None of the time" to "All of the time" and are asked to select the option that describes them the best.A higher total score represents a higher level of hope.Snyder et al. (1997) has reported acceptable psychometric properties such as internal consistency (r = 0.77) and test-retest reliability (r = 0.73), as well as support for Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) concurrent and predictive validity.A reliability index of α=.787 was found (Cronbach´s Alfa) in this study.
Coping Skills Questionnaire (Hernández-Guzmán, 2003) was developed and standardized in México to assess coping skills in children.The Cuestionario de Afrontamiento is a self-report measure for children aged 6 to 12 years.The scale has 12 items related to child´s interpretation and reactions when facing a problem, and the things he or she does to cope and/or solve the problem.Lower scores reflect a more proactive positive coping.Children are asked to rate, on a 3-point scale ranging from never (0) to always (2), the frequency with which they experience each statement.

Capacidades y Dificultades is the Spanish version of the Strengths and Difficulties
Questionnaire.The SDQ-S is a 25 item self-report measure of psychological adjustment for use with children aged 11 to 17 years.The SDQ-P is completed also by parents or caregivers to evaluate psychosocial adjustment of children and adolescents aged 3 to 16 years.The items are divided between five scales: emotional symptoms, conduct problems, hyperactivity/inattention, peer relationship problems, and pro-social behaviour.Participants are required to endorse either "not true" (0), "somewhat true" (1), or "certainly true" (2) in response to each statement, with higher scores indicative of greater problems for each subscale except for pro-social behaviour.The SDQ -S has sound psychometric properties, including moderate to strong internal reliability for all subscales, good test-retest reliability (Vostanis, 2006), concurrent validity, and the ability to distinguish between community and clinical samples (Goodman, 2001).The measure has not been validated with Mexican population.However the validation study from Spain has shown acceptable psychometric properties: Cronbach's alpha coefficient for the total SDQ for students range between .77and .87,and the SDQ for parents/caregivers between .76 and .83(Rodríguez-Hernández, 2006;Rodríguez-Hernández et al., 2012).A reliability index of α=.731 was found (Cronbach´s Alfa) in this study.
Social Validity Questionnaire for Children and Adolescents (Barrett, 2005).For this study, the questionnaire was translated into Spanish.The questionnaire is comprised of seven questions.Using a 4-point scale from 4 (a lot/all the time) to 1 (not at all/ nothing at all).The first five questions related to how enjoyable the program was, how much they learned by doing the program with their friends, how much they learned about feelings and about how to cope when being worried or upset, and how often they use the skills.The sixth question asked which skills did they found most useful, and a final open-ended question provide an opportunity to comment on the program.
Social Validity Questionnaire for Parents/Caregivers (Barrett, 2005).For this study, the questionnaire was translated into Spanish.The questionnaire was comprised of eleven questions.The first nine questions used 4-point scale from 4 (a lot/all the time/very useful) to 1 (not at all/ nothing at all/not useful at all).Two of the questions asked how useful they thought positive coping-skills programs were and how important will it be to incorporate a program like this into the school curriculum.Two questions related to how useful the skills were to themselves and how much did they learn about enhancing their child/adolescent´s coping skills.Four questions related to how much they thought their child/adolescent learnt about feelings and coping skills, how much they thought she or he enjoyed the program, and how often did she or he used the ideas and skills learnt in the program.Question 10 asked parents/caregivers to choose which skills their child/adolescent found more useful and question 11 was an open-ended asking to provide any other feedback (positive or negative) that they had.

Procedure
Measures were counterbalanced and administered to all participants at pretest and posttest.Specifically, pretest measures were administered three days before implementing the program and posttest measures three days after the program finished.
Instructions and test items for all measures were read aloud and participants were informed that all responses were confidential.

Intervention protocol and materials
The acronym is parallel to the English in terms of the concepts taught.After the introductory session, children start to learn the letter F, which stands for "Feeling worried?" followed by the letter R "Relax and feel good", I "Inner helpful thoughts", E "Explore solutions and coping plans", N "Nice work; reward yourself", D "Don´t forget to practice" and S "Smile and stay calm".Within each session, the teacher uses modeling of the skills, and after the skills are taught, children have opportunities to practice in small groups and debrief with the whole classroom.The program has three developmentally appropriate versions in Spanish, each of which were used in the current study: AMISTAD y Diversión for children ages 4-7 (Barrett, 2011a;Barrett, 2011b), AMISTAD para Siempre for children ages 8-11 (Barrett, 2011d;Barrett, 2011e), and Programa de Resiliencia para Adolescentes for adolescents ages 12-16 (Barrett, 2011d;Barrett, 2011f).

Statistical Analysis
Questions were addressed separately using paired sample t-tests (alpha level 0.05) in order to examine the dependent variables of anxiety, self-concept, hope, coping skills, and psychosocial difficulties.Means, standard deviations, and frequencies were calculated for each item of the social validity questionnaires.

Effects of the Program
To examine the effect of the intervention, participants´ total scores and subscales on each outcome measure were compared from pretest to posttest.Table 1 displays the means and standard deviations on each dependent measure at each time point as well as t-test significance pre-post intervention.
The results showed that all the subscales of the anxiety measure, CMAS-R, as well as the total score, reported a statistically significant decrease after receiving the program.A statistically significant improvement was also found on girls´ proactive copings skills.In the same way, the Strengths and Difficulties Questionnaire answered by caregivers also reported statistically significant improvements.The subscales of Physical Appearance and Attributes and Popularity of the Piers Harris Children´s Self-Concept Scale 2 also reported a statistically significant increased.

Social Validity Evaluation
Girls and caregivers also completed a social validity evaluation with the purpose of knowing their opinion about how enjoyable and useful the program was.Table 3 displays the girls´ responses to questions regarding how much they enjoyed the program, how much they learned, and how often they practice the ideas and skills learned in the program.Overall, girls evaluated the program as enjoyable and helpful to cope with difficult situations; 94% indicated that they enjoyed the program and 88.7% reported that they used the skills learned in the program.
Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) Frequencies and percentages of the skills that girls found most useful appear in Table 4.The two coping skills that they found most useful were "relaxation exercises" (84.6%) and "helping others feel good" (82.7%).The two skills taught that they found least useful (although still relatively useful) were "deep breathing" (71.2%) and "the sixblock problem solving plan" (73.1%).
Table 5 displays the caregivers´ responses on each of the items of the social validity questionnaire.Overall, they reported that the program was useful for them and for their children/adolescents; 96% indicated that they learned about enhancing their child/adolescent´s coping skills and 80% reported that the program was useful for this.
Frequencies and percentages of the skills caregivers found most useful for their children/adolescents appear in Table 6.The two coping skills that they found most useful were "changing negative thoughts to positive thoughts" (81.1%) and "recognizing feelings in yourself" (75.7%).The two skills taught they found least useful were the "sixblock problem solving plan" (32.4%) and "relaxation exercises" (36.8%).Results from Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) 14 the social validity questionnaires suggested that the program was, in general, well received by the girls living in the orphanage, and the caregivers.

Discussion
Being institutionalized in an orphanage has been linked to a series of negative psychological and social consequences (González et al., 2012) such as: the development of insecure attachment, lack of impulse control, low self-esteem, conflict in interpersonal relationships (Martínez, Rosete, & de los Ríos, 2007), a higher external locus of control, lower perception of self-competency, higher perception of family difficulties, and a negative and avoidant coping style (González, Valdez, Domínguez, Palomar, & González, 2008).Despite these elevated risks for negative outcomes, there  et al., 2004;Siu, 2007;Stopa et al., 2010).Several of the core components of the FRIENDS program include anxiety reduction techniques such as relaxation exercises, cognitive restructuring, and in vivo gradual behavioral exposure.
Therefore, learning these techniques may have helped them to decrease their anxious symptoms.This idea is also supported by the social validity results, as relaxation exercises was rated by girls at the most useful skill learnt in the program.
Improvements were also observed in the Piers Harris subscales of Physical Appearance and Attributes, and Popularity.Other studies have reported changes in similar areas.For example, when implementing the FRIENDS program with children who had recently immigrated to Australia, a couple studies reported improvement in participants´ levels of self-esteem and reduced peer victimization (Barrett et al., 2001;Barrett et al., 2003).In the same way, current results as similar to those found by Gallegos et al. (2012), who reported that after receiving the program, their participants´ reported improvements in the Home and Social Self-esteem subscales of the Coppersmith Self-Esteem Inventory.These improvements are in line with the components of the FRIENDS program that target learning about empathy, self-respect and assertiveness, and social skills that promote positive relationships.
Girls from the current study, after receiving the program, reported statistically significant increases in their proactive coping skills, as they were better able to cope with anxiety-provoking situations and reduced their avoidance behaviors.This is similar to what has been found in other studies such as the ones conducted by Gallegos et al., (2012) and Stopa et al. (2010).This may be the result of learning different coping strategies such as positive thinking, relaxation and gradual exposure coping plans.
Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) Caregivers reported reductions in girls´ emotional and conduct problems and higher levels of pro social behavior.Interestingly, both caregivers and girls reported that "helping others to feel good" was one of the most useful skills they learned.This may be a reflection of culture, as socializing, cooperating and helping others are highly value in Latin-American societies.Therefore, incorporating altruistic activities within their regular routines could also be potentially beneficial to increase children´s own happiness and wellbeing.
Contrary to our original hypothesis, and to the findings of Barrett, Sonderegger, and Xenos (2003), the current study did not find differences in the levels of optimism.A possible explanation could be that measure was not sensitive enough to evaluate optimism in children being reared in Mexican culture, as it had not been previously validated in this population.It could also be the case that, due to the specific characteristics of our participants, the program would need to be adapted in a more culture-specific and contextually relevant way, for those children living in an orphanage setting.These children may be dealing with significant issues of child traumatic stress and grief and loss in order to create movement on this variable of optimism and hope.
Thus, future research should examine the impact of refined contextualization of the program.
Also contrary to hypothesis, girls´ scores on the SDQ did not indicate statistically significant changes.One possible explanation could be a desirability effect, in which the girls participating in the study could have elevated their evaluations of themselves at baseline, supported by the idea that children and youth in institutionalized care may have an increased need for approval (González et al., 2008).

Limitations and Future Directions
Although this study has included a larger sample size (n=57) than the previous study (Gallegos et al., 2012), the sample size still limits the power to reach statistical In terms of the design of the study, the current study lacks a control/comparison group which was precluded in the current study due to time and financial constraints, and to the particular characteristics of the setting.Because of this, a potential threat to internal validity exists because we cannot confirm if the difference between pretest and posttest scores is related to the intervention or to the maturation and history of participants.Threats to construct validity could have occurred such as the "monomethod bias", as for all variables self-reported measures were used.Also novelty of the intervention could have prompt participants to respond unusually well to the program (Shadish, Cook & Campbell, 2002).
It is recommended that a larger study be conducted with a control/comparison condition.While the current study included caregivers as additional informants, it would be additive to also include teacher reports in future studies as well as to evaluate the effectiveness of program for both genders.Neil and Christensen (2009) have suggested that the biggest effect sizes for anxiety prevention and early intervention programs are seen for high-risk groups.
Clearly then, young people living in orphanages both 1) have an urgent need to improve adaptive social-emotional skills and increase their resilience and 2) best able to recognize substantial benefits from such interventions.Resilience has been related to a number of different positive outcomes such as decreasing the risk for experiencing psychopathology, better physical and mental health, and higher levels of happiness and subjective wellbeing (Friborg, Hjemdal, Rosenvinge, & Martinussen, 2003).
Particularly for children and youth living in an orphanage, a high level of resilience seems to be almost a requirement to thrive (Ahmad & Mohamad, 1996;Somen, 1986).Future studies should delve further into the social and emotional development of these vulnerable children and youth in order to identify which the specific mechanisms will increase their resilience and assist in providing the most enriching, stable and nurturing environment possible.In order to produce a meaningful and sustainable change, one important direction should include an emphasis on building resilience and emotional intelligence in caregivers.This may be particularly important for young people living in an orphanage as they may not have had positive adult role models and the adults in their lives can serve as critical change agents for Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar)The six group leaders that implemented the intervention completed a two-day training covering the principles and practices of prevention and early intervention and a step-by-step guide to implement the FRIENDS program.The group leaders had a Bachelor in Psychology, and their age range was between 25 and 30 years old.Eight groups were randomly formed taking into account age and each group was comprised of six to eight participants.There are three developmentally appropriate versions of the FRIENDS program.One developed for children from 4 to 7 years old, one for 8 to 11 years old, and one for adolescents from 12 to 16 years old.The groups were formed for each version of the program.When they were more than eight participants for a specific version, the group was divided to include participants from a closer age range (i.e., 10-11 years old).Group leaders implemented the program once a week for 10 consecutive weeks.Sessions lasted 60 to 75 minutes and were conducted after school at the orphanage.
FRIENDS program was developed by Dr. Paula Barrett with the aim of increasing adaptive social and emotional skills, promoting resilience and preventing anxiety in children and youth.The program is based in cognitive-behavioral theory (CBT) and positive psychology.Based on efficacy research, CBT is now recommended as the gold standard for treating and preventing anxiety (Neil & Christensen, 2009).It incorporates physiological, cognitive, and behavioral strategies to assist children in coping with stress and worry.The behavioral component includes self-monitoring of feelings and thoughts, out-of-session and mental imagery exposure and relaxation training.The cognitive component teaches children to recognize their feelings and thoughts and the link between them.It also teaches students to identify faulty cognitions and incompatible self-statements, and to elaborate alternative interpretations for difficult situations.The family and peer support component discourages the avoidance of anxiety-provoking situations by promoting the practice of problem solving.It encourages the building of social support groups and respect for diversity.Learning techniques include group discussion, hands-on activities, and role-play.Approximately, one session is dedicated to learn each of the seven steps represented by the FRIENDS acronym.The Spanish Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) has been a dearth of research related to preventive interventions for anxiety and psychosocial difficulties among children living in orphanages, particularly in Latin American countries such as Mexico.This study investigated the effectiveness of the Spanish version of the FRIENDS program, used as a selective prevention strategy with girls living in an orphanage.The purpose was to examine whether the program produced any changes in anxious symptoms and psychosocial difficulties as well as changes in the protective factors of Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar) self-concept, hope, coping skills and pro social behavior.Satisfaction with the program was also assessed.A positive and statistically significant change was found for anxiety, with girls reporting lower levels of anxious symptomatology after receiving the program.This is similar to what has been found with other "at risk" population such as: immigrants in Australia, African-American youth exposed to violence, low SES children and youth in the United States, and Chinese children identified at risk for developing internalizing disorders (Cooley-Quille significance and future research would benefit from larger sample sizes within a similar population.It also should be taken into account the low reliability indices reported for two of the measures: CMAS-R and Coping Skills Questionnaire.Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar)

Table 1
Means and SD for the R-CMAS, Self-Concept, Hope, Coping Skills and SDQ at Each Time Point * p<.05 , ** p<-01 Pensando Psicología / Volumen 10 / Número 18 / enero-diciembre 2015 (Artículo sin editar)Pair samples t-tests were also conducted on the items of each outcome measure.The items that reported statistically significant differences were the following.From the Piers Harris Children´s Self-Concept Scale 2, item 30 "My parents expect a lot from me", t (41) = -2.72,p <.05.From the Children´s Hope Scale, the item 6 "Even when others Note. 4 A lot / All the time, 1 Not at All/ Nothing at All.

Table 3
Participants´ ratings of the skills they found most useful.

Table 4
Caregivers´ means for each of the items of the Social Validity Questionnaire.

Table 5
Caregivers´ ratings of the skills their child found most useful.