Development and Validation of the Arab Youth Mental Health Scale

Community Ment Health J (2011) 47:331–340 DOI 10.1007/s10597-010-9312-6 ORIGINAL PAPER Development and Validation of the Arab Youth Mental Health Scale Jihad Makhoul • Rima T. Nakkash • Taghreed El Hajj • Sawsan Abdulrahim • Mayada Kanj • Ziyad Mahfoud • Rema A. Afifi Received: 6 September 2009 / Accepted: 12 April 2010 / Published online: 6 May 2010 Ó Springer Science+Business Media, LLC 2010 Abstract A variety of measures of mental health have been used with youth. The reason for choosing one scale over another in any given situation is rarely stated, and cross-cultural validation is scarce. Psychometric testing is crucial before utilizing any measure of mental health with a certain population, due to possible cultural variations in interpreting meaning. The research reported herein describes the development and psychometric testing of the Arab Youth Mental Health Scale. The process included 5 phases: (1) reviewing existing scales leading to the identification of 14 non-clinical and relatively short mental health scales used previously with youth; (2) rating the scales by the researchers and community members leading to the identification of 3 scales with apt structure, and that were judged to be suitable, applicable, and appropriate; (3) soliciting youth input to assess comprehension of each item in the selected 3 scales and to discover context specific mental health related feelings, thoughts, and expressions; (4) seeking expert opinion to classify items remaining after phase 3 that measured common mental disorders, and to limit repetitiveness; and (5) testing for psychometric properties of the 28 items that remained after the previous 4 phases. The contribution of each phase to the process is described separately. Results of the exploratory principal J. Makhoul
R. T. Nakkash
T. El Hajj
S. Abdulrahim
M. Kanj
R. A. Afifi (&) Department of Health Behavior and Education, Faculty of Health Sciences, American University of Beirut, P.O. Box 11-0236, Riad El Solh, 1107 2020 Beirut, Lebanon e-mail: [email protected] Z. Mahfoud Department of Epidemiology and Population Health, Faculty of Health Sciences, American University of Beirut, P.O. Box 11-0236, Riad El Solh, 1107 2020 Beirut, Lebanon component analysis resulted in one factor which explained 28% of the variance and for which 21 items loaded above an eigenvalue of 0.5. No other factor added significantly to the explanation of variance, nor had items that added theoretical or conceptual constructs. The process of soliciting feedback from youth groups, the community and professionals; and of field testing was challenging; but resulted in a contextually sensitive, culturally appropriate and reliable scale to measure mental health of youth. We recommend that researchers measuring mental health of youth critically analyze the relevance of existing scales to their context; consider using the AYMH scale if appropriate to their target population; and when needed, use a similar methodology to construct a relevant, culturally and contextually sensitive measure. Keywords Youth Mental health
Validation
Arab
Introduction Approximately 1 in 4 or 5 adolescents will suffer from a mental health problem in any year (Patel et al. 2007). A variety of risk and protective factors have been linked to the probability of mental health problems, with refugee status a definite risk. (Patel et al. 2007). A variety of measures of mental health have been used with youth generally (Harpham et al. 2003; Myers and Winters 2002b; Winters et al. 2002; Costello and Angold 1998), and specifically with refugee youth (Keyes 2000; Mollica et al. 1997; Nese et al. 2005; Rousseau and Drapeau 1998; Savin et al. 1996; Slodniak 2002). However, most current measures of mental health are used for clinical diagnosis at the individual level rather than for assessing prevalence at the 123 332 population level (Myers and Winters 2002b). Additionally, despite the abundance of youth mental health scales in the literature, the reason for choosing one scale over another in any given situation is rarely stated, and cross cultural validation is scarce. Psychometric testing is crucial prior to utilizing any measure of mental health with a certain population, due to possible cultural variations in interpreting meaning (Moreau et al. 2009; Hundt et al. 2004; Ommeren 2003). Palestinians were displaced beginning in 1948 and became refugees residing in Lebanon, Syria, and Jordan. An estimated 422,000 Palestinian registered refugees reside in Lebanon (UNRWA 2009). Palestinian refugees in Lebanon live under dire environmental and social conditions. These conditions are commonly perceived to be the worst of Palestinian refugees in the region, due to limited employment opportunities, scarce economic resources, and limited access to basic health and social services—exacerbated as a result of state imposed restrictions on employment and opportunities to seek education (Jacobsen 2000). Health and social services are provided by a variety of international as well as governmental and non-governmental organizations. The United Nations Refugee and Works Agency (UNRWA) was set up in 1948 specifically to provide educational and health services to the Palestinian refugees. However, services are fragmented and insufficient. Palestinian refugees in Lebanon have the highest prevalence of mental distress when compared with other such refugees in the region (Jacobsen 2000). Using an abbreviated version of the Hopkins Symptoms Checklist, refugees aged 15 years and older in Lebanon were found to have a mean score over 3 (on a scale of 0–7), the highest mean score as compared to refugees in Syria and Jordan (Jacobsen 2000). A particular study of relevance to this population involved 590 never married 13–19 year old Palestinian refugees (mean age: 15.84, SD: 2.02) living in Burj El Barajneh camp (BBC) in Beirut (Makhoul and Nakkash 2009). BBC—located in the southern suburbs of Beirut, is the 6th largest of the 12 official camps established in Lebanon to house Palestinian refugees after 1948. BBC houses approximately 14,000–18,000 residents over an area of 1.6 square kilometers (Statistics UNRWA 2009; Makhoul 2003). Though mental health was not specifically measured, several potential indicators were evident. For example, youth surveyed lived in households where the mean annual income of the household was 4,854,000L.L. (LL1500 = $1.00). Forty five percent of the youth surveyed were out of any educational institution (school, university, technical). Over a quarter (27%) of the youth were working. Of those who worked, 67% worked more than 40 h per week, and 90% earned less than 75,000L.L. per week. In addition, 63% contributed at least part of their 123 Community Ment Health J (2011) 47:331–340 income to household expenses. Approximately 15% of the youth surveyed reported they had never been to see a doctor, and 24.1% had seen a doctor over 1 year ago. The youth had been exposed to a variety of stressful life events: 42% had had a death in the family in the past year, 67% had a family member hospitalized, and 48% had a parent who had taken a loan and had to repay it. When asked to compare themselves to others their age in Lebanon, 68% stated that they have fewer opportunities compared to others. In addition, only 13% stated that they felt their life would improve a lot in the future. With respect to social capital, 94% of those surveyed trusted few people or no one in their neighborhood and 67% stated that one needed to be vigilant and cautious when dealing with neighbors. Finally, 59% had not exchanged a favor with a family member, 76% had not done so with a friend, and 80% with a neighbor (Afifi et al. 2010). With specific relevance to Arab youth, a variety of mental health scales have been used among Arab refugee youth (Bean et al. 2007; Foldspang and Montgomery 2000), and Arab youth generally, (Baker and Kanan 2003; Elbedour et al. 2007; Giacaman et al. 2004; Hundt et al. 2004; Punamaki et al. 2005; Thabet et al. 2000, 2002; Thabet and Vostanis 1998, 1999, 2001). These include the Hopkins Symptoms 37 (Bean et al. 2007), the Child Post Traumatic Stress Reaction Index (CPTSD-RI; Baker and Kanan 2003; Thabet et al. 2002; Thabet and Vostanis 1999), the Revised Child Manifest anxiety scale (RCMAS; Thabet et al. 2002; Thabet and Vostanis 1998), the Rutter scale (completed by teachers about children; Thabet and Vostanis 1998, 1999, 2001), the Gaza traumatic events checklist (Thabet and Vostanis 1999), and the Children’s Depression Index (Arabic version; Baker and Kanan 2003) among others. However, despite the abundance of measures used to measure Arab youth mental health, psychometric properties of these instruments are rarely measured (Moreau et al. 2009; Abdel-Khalek 2002; Foldspang and Montgomery 2000). In an effort to address the dearth of robust analysis of scales used with Arab adolescents, this paper describes the process of development and psychometric testing of the Arab Youth Mental Health (AYMH) Scale. The AYMH scale was developed to measure the impact of an intervention planned as a follow-up to the survey conducted in BBC and described above. The results of that survey were disseminated at a meeting of stakeholders (NGOs, UNRWA, and adult and youth residents of BBC). At this meeting, a decision was taken to move towards intervention to promote mental health of youth in BBC, and a Community Youth Committee (CYC) was established to guide the development, implementation and evaluation of an intervention using Community Based Participatory Research as a conceptual framework (Viswanathan et al. 2004). Community Ment Health J (2011) 47:331–340 Methods The process of developing the Arab Youth Mental Health (AYMH) Scale included 5 phases (Fig. 1): (1) reviewing existing scales; (2) rating the scales by the researchers and community members; (3) soliciting youth input; (4) seeking expert opinion; and (5) testing for psychometric properties. Each is described separately below. The phased approach to development and validation of instruments is supported by previous research on childhood autism 333 (Schopler et al. 1980), job stress (Spector and Jex 1998, 2003), and mental well being (Tennant et al. 2007). Reviewing Existing Scales The researchers conducted a review of existing mental health scales used with youth from the published literature using medline, pubmed, PsycINFO, academic search premier and Google scholar. Additional scales were identified through an iterative process of reading published articles Fig. 1 The 5 phases of development and validation of the Arab Youth Mental Health Scale 123 334 and reports on mental health measurement, mental health interventions, and refugee health. Throughout the search process, the researchers sought to review scales that were non-diagnostic and relatively short, keeping in mind that these will be used with youth in community settings. Search terms included mental health, anxiety, depression, and quality of life—paired with adolescent or youth. The inclusion criteria comprised scales measuring depression and anxiety as well as other scales measuring quality of life pertinent to mental health. Exclusion criteria included scales used for diagnosis only (never used in a community sample or for screening) and that were long (over 60 items unless used specifically for screening and in refugee populations). Most of the scales identified had been used either in clinical settings for screening or in population-wide surveys to assess prevalence of mental health while a smaller number of scales had been used specifically to evaluate impact of interventions. The search yielded 14 scales for review. These were: the Self Report Questionnaire (SRQ20; Harpham et al. 2003; Harding et al. 1980), the Hopkins Symptoms Checklist 25 (HSC; Mattsson et al. 1969), the WHO Quality of life-Bref (WHO QOL; WHOQOL 1998), the Harvard Trauma Questionnaire (HTQ; Mollica et al. 1992), the Short Form 36 Question (SF36; Ware and Sherbourne 1992), the Mental Health Inventory-5 (MHI-5; Veit and Ware 1983), the Community-based psycho social support survey questionnaire used by Birzeit University Institute of Community and Public Health (CPSSQ; Giacaman 2004), the Duke Health Profile (DHP; Parkerson et al. 1990), the Affect Balance scale (ABS; Bradburn 1969), the Center of Epidemiological Studies-Depression score (CES-D; Radloff 1977), the General Health Questionnaire (GHQ-12; Goldberg and Hiller 1979), the Kessler-6&10 (K-6 or K-10; Kessler and Mroczek 1994), the Mental Health Inventory from the Medical Outcomes Study (MHI-MOS; Ware et al. 1992) and the Strength and Difficulties Questionnaire (SDQ; Vostanis 2006). Rating of the Scales by the Researchers and Community Members To select the scales which were most relevant to the local context, the researchers sought input from members of the Community Youth Committee (CYC). As mentioned above, the CYC was established to guide all phases of the intervention project, and included 17 different NGOs that work with youth in the camp, camp residents including youth, UNRWA, as well as academicians from the American University of Beirut. Usually each NGO sent one representative that consistently attended the meetings. For the purpose of reviewing the scales, a subcommittee was formed, since a smaller group was thought to facilitate 123 Community Ment Health J (2011) 47:331–340 discussion and make the process more efficient. The subcommittee consisted of two members from the academic research team and three volunteer members from the CYC (non academic). The process was very time consuming. The subcommittee met on average 5 times to review the selected 14 scales and each meeting lasted up to 2 h. At the end of each meeting, the research team members summarised discussions to inform the next meeting. The scales were rated according to set criteria as described below. Each member of the committee discussed why they thought a certain scale fit the criteria for selection or not and a discussion ensued. Some scales took more time for discussion than others to reach a consensus. The purpose of this phase was only to select those scales most relevant to our context rather than adapt them—which was the purpose of phases that followed. Members of the subcommittee looked over the 14 scales and rated each using a set of criteria based on the reviewed literature (Bowling 2001, 2005; Boyle and Jones 1985; Myers and Winters 2002a). The criteria included: Suitability, the extent to which the instrument items were culturally relevant for use within the camp; Applicability, the extent to which the youth would understand the questions; Structure, the extent to which the recall period, the number of items in the questionnaire and type of responses (Likert or dichotomous) matched the norms of the context; and Appropriateness, or relevance to the developmental stage of youth. Since none of these particular scales had been tested for validity and reliability in this specific context and with the population in question, we discounted this as a relevant criterion in the selection process. Table 1 summarizes the ratings for each scale. Structure was considered first as it was the most straightforward to assess. Based on their previous research experience with this age group in comparable contexts, the researchers thought that measures with long recall periods were harder to use with 10–14 year olds because of difficulty with recall. Although Myers and Winters (2002a) maintains that a dichotomous response format is better for use with youth, it was apparent from prior experience that a yes/no option did not work very well with youth in this context. Dichotomous response options were felt to result in underestimation of mental health issues as the youth would tend to answer in the negative to avoid being ‘labeled’. Also based on prior experience with this age group in a comparable context, a short Likert type format was found to be more comprehensible. Consequently, the SRQ20, GHQ12, MHI-MOS, CPSSQ, ABS, Kessler, HTQ, DHP, WHO-Quality of Life, WHO MHI-5 and SF36 were all excluded (see Table 1 for details). Once the list of 14 scales had been shortened to 3, the researchers and coalition members reviewed the remaining scales using the Community Ment Health J (2011) 47:331–340 335 Table 1 Rating of scales according to set criteria Scale Structure Suitable Applicable Appropriate SRQ 20 No (Dichotomous; Recall: 1 month; 20 Q) No Yes No Hopkins checklist Yes (Not dichotomous; Recall: 1 week; 25 Q) Yes Yes Yes WHO Quality of life-Bref No (Not dichotomous; Response categories too wide; Recall: 1 month; 26 Q) Yes Yes No Harvard Trauma Q No (Dichotomous; Many open ended Q’s; Recall: Ever; 92 Q) No Yes No SF 36 Maybe (Some dichotomous; Response Options vary across Q’s; Recall: *1 month; 36 Q) Yes No No WHO MHI-5 No (Not dichotomous; Recall: 1 month; 5 Q) Yes Yes Yes Community-based psychosocial support survey No (Some open ended q; Response Options vary across Q’s; Recall: 2 weeks–6 months; 43 Q) Yes Yes Yes Duke health profile Maybe (Some dichotomous; Response Options vary across Q’s; Recall: 1 week; 17Q) Yes No Yes Affect balance scale No (Dichotomous; Recall: past few weeks; 10 Q) No Yes No CES-D GHQ-12 Yes (Not dichotomous; Recall: 1 week; 20 Q) No (Not dichotomous; Categories difficult for youth; Recall: past few weeks; 12 Q) Yes Yes Yes No Yes No Kessler 6 or 10 No (Not dichotomous; Response categories too wide; Recall: 1 month; 6–10 Q) No Yes No Mental health inventory from the medical outcomes study Strengths and difficulties No (Not dichotomous; Response categories too wide; Recall: 1 month, 57 Q) Yes (Not dichotomous; Recall: 6 months; 33 Q No Yes No Yes Yes Yes other three criteria (applicability, suitability, and appropriateness) which were considered in tandem. The three remaining scales were: the SDQ, the Hopkins and the CESD. The SDQ has non-dichotomous response items that were similar for all the survey items (except for 8 clinical items included in the SDQ but not considered for our purposes). Although it has a long recall period (6 months), it was found otherwise suitable, applicable, and appropriate. The Hopkins checklist has non-dichotomous response options, only 25 items, and a recall period of 1 week. One question related to sexual pleasure seemed inappropriate, but with this item disregarded, the Hopkins checklist was appropriate, applicable, and suitable. The CES-D has a 1 week recall period, is not dichotomous, has scale options that were thought to be easy and comprehensible by youth and has only 20 questions in total. The CES-D was also found to be appropriate, applicable, and suitable for our population. These three scales were subsequently translated into formal Arabic to be used in subsequent phases. Soliciting Youth Input To check for comprehension and the appropriateness of terms used in the scales to express mental health and related symptoms, the researchers asked the opinions of a sample of out-of-school youth (drop-outs) in the camp through two focus groups. Participatory research particularly involving youth suggests that by listening to young people’s stories and collaborating with them in designing interventions to address their concerns, research questions and interventions are more effective (McIntyre 2000). This is because the youth are particularly well aware of issues that affect them and they will benefit more from programs they participate in setting up (O’Donnell et al. 1997). The researchers were committed to hearing youth voices. Thirteen youth aged 13–17 years accessed through Palestinian NGOs in the camp serving out-of-school youth were invited to participate in focus group discussions conducted in the camp by two of the authors (JM, RN). The youth found the items in formal Arabic difficult to understand. The researchers felt this would also apply to youth still in school as well. Thus, the scale needed to be re-edited into colloquial Arabic which all children 10–14 years old could understand. Further FGD were conducted at the University by 4 of the authors (JM, RN, RA, MK) with 44 youth in grades 5 and 6 from UNRWA1 schools in the camp. The focus groups checked for: (1) comprehension (2) relevance (3) response options, and (4) context specific feelings, thoughts, and expressions. The findings revealed that several of the questions were conceptually meaningless to the young persons and many terms were ambiguous to them. The double barreled items containing two questions in one, such as ‘‘I fight a 1 UNRWA is the United Nations agency established for the sole purpose of providing educational and health services to Palestinian refugees. 123 336 lot. I can make other people do what I want’’ appearing in the Strength and Difficulties Questionnaire is one such example. Also, youth found difficulty with the recall period of ‘during the past 6 months’. Instead, they found it easier to recall seasons (winter or summer) as most of their year is spent either in or out of school; or significant events (the summer war), or the number of the month (‘‘month 6’’ meaning June). In addition, the young people stated that they remembered events most clearly in the last week. They also seemed to have difficulty understanding the slight difference between the 4 response options of the CES-D (rarely or none of the time, some or little of the time, occasionally or a moderate amount of the time, and most or all of the time), and the 4 response options of the Hopkins Symptom Checklist (Not at all, a little, quite a bit, extremely), and the 3 response options of the SDQ (not true, somewhat true, certainly true). Using stars to indicate increasing intensity of feeling (or frequency) was the preferred alternate option among the participants. The youth used specific Arabic terms to express how they felt about recent incidents that affected them the most, such as the Israeli attack on Lebanon in the summer of 2006 and other matters which continue to affect them, such as having no place to play, confinement to the camp, crowding, noise and violence. The rich information from these group discussions provided the researchers with a list of terms that were sure to capture the youth’s experiences. The most recurring terms the boys used were: ‘was sad’, ‘was afraid’, ‘cried’. Examples of other terms mentioned at least once to express their feelings were: ‘devastated’, ‘felt sorry’, felt lonely’, ‘was bored’, ‘couldn’t eat’, ‘lost my temper’. A number of them spoke of nightmares and thoughts of death and dying. Similarly, terms such as ‘afraid’, ‘irritated’, ‘agitated’, ‘upset’, ‘depressed’, recurred in the focus groups with girls and other terms that were mentioned at least once included: ‘pressured’, ‘couldn’t concentrate’, ‘dizzy’, ‘lost hope in life’, ‘started to shake’, ‘sadness’, ‘worn out emotional state’. They, too, spoke about death and dying, nightmares and being disturbed by the war and the violence in their neighborhoods. They also expressed their fear for their loved ones, but let out their feelings by screaming and talking to their friends about their problems. The researchers then formulated statements using these terms or chose statements that were part of the three short listed translated questionnaires. A new scale of 40 items; a shorter Likert scale with three options: always, sometimes and rarely; and a shorter recall period (1 week) was formulated. Seeking Expert Opinion To contextualize the resulting scale further and to point out the questions which were the most relevant to use in screening for common mental disorders, the researchers 123 Community Ment Health J (2011) 47:331–340 sought the input of four local mental health specialists, two psychologists and two psychiatrists. This constituted the content validation aspect of the scale. They each responded with suggestions about questions to keep and to delete. The items which they all agreed to keep remained in and those which they all deleted were removed. The remaining items were then reviewed and kept if two out of four experts suggested including them. This round of expert comments resulted in a scale of 28 items. Testing Psychometric Properties To test for construct validity of the latest draft of the mental health scale (28 items), the researchers administered the questionnaire to a total of 288 students in fifth and sixth grades in UNRWA schools. The data were entered into SPSS version 15 for analysis. First, an exploratory factor analysis was carried out on the 28 items utilizing the principal components extraction (PCA) method. We used PCA to identify a more parsimonious set of items, or the smallest number of factors, that could explain most of the variance in the data (Bryant and Yarnold 2000). We used PCA because our intent was to carry out exploratory, as opposed to confirmatory, factor analysis. Secondly, we followed the factor analytic step with ANOVA tests for the purpose of construct validity. It was hypothesized that youth who score low on the mental health scale (i.e., express a higher level of psychological/mental distress) will also report poor general self-rated health (GSRH) and poor self-rated mental health (SRMH). Results Based on the exploratory factor analysis using the principal components analysis (PCA), we got 7 possible factors with total eigenvalues over 1 (Table 2). Factor 1 explained 28.319% of the total variance in the data. None of the other factors contributed substantially to explaining more variance. Three of these seven factors (factors 1, 2, 7) had items with loadings of 0.5 or above (Table 3). The cut-off point of 0.5 was chosen because our intent was exploratory and in order to include as many items as possible. However, upon closer examination, most of the items either loaded or did not load on factor 1. There were two items that loaded only on factor 2 (feeling comfortable and secure wherever I went; feeling happy), one that loaded only on factor 7 (feeling that nothing mattered, not caring about anything), and four that did not load well on any of the factors (feeling bothered by things that usually do not bother me; shivering without being cold; losing appetite; and wanting to hit someone). The three items that loaded on factors 2 and Community Ment Health J (2011) 47:331–340 337 Table 2 Results of the principal components analysis—factors and variance explained Total variance explained Component Initial eigenvalues Total % Of variance 1 7.929 28.319 2 1.727 6.168 3 4 1.436 1.267 5.128 4.525 5 1.191 6 1.046 7 1.027 Extraction sums of squared loadings Cumulative % Total % Of variance Cumulative % 28.319 7.929 28.319 28.319 34.487 1.727 6.168 34.487 39.615 44.140 1.436 1.267 5.128 4.525 39.615 44.140 4.252 48.392 1.191 4.252 48.392 3.737 52.129 1.046 3.737 52.129 3.669 55.798 1.027 3.669 55.798 Extraction method: principal component analysis Table 3 Results of the principal component analysis—factor loadings for each item within identified components Component matrixa Component 1 2 3 4 5 6 7 During the last week I was upset .562 .228 -.003 -.373 -.072 .354 -.119 During the last week I burst into tears several times .520 .370 -.085 -.101 -.167 .032 During the last week I was bothered by things that usually do not bother me** During the last week I was feeling scared and frightened .432 -.142 .520 .206 During the last week I felt suffocated .663 .120 During the last week my sleep was interrupted because I was thinking of so many things .641 .057 -.101 .099 -.209 -.092 -.124 During the last week I was shivering without really being cold** .398 .225 -.453 .015 During the last week I was tense/nervous .608 -.239 .116 -.007 -.185 -.175 -.090 During the last week I felt lonely .561 -.009 .168 -.369 -.068 -.350 -.112 During the last week I lost my appetite** .448 -.024 .059 -.467 .057 .235 -.140 -.400 -.043 -.219 -.342 .289 -.153 .086 -.132 .137 -.015 -.100 -.160 .170 -.429 .043 .091 .148 .238 .335 .066 -.067 .096 During the last week I was sad .618 .175 -.071 -.245 During the last week I was worried .644 .155 -.063 .216 -.173 During the last week I was having difficulty concentrating on what I was doing .526 .105 --.041 .374 -.137 During the last week I felt dizzy/light headed .554 -.318 -.208 -.160 .316 During the last week I didn’t feel like talking .621 -.020 -.047 -.191 .014 -.320 .301 .109 -.052 -.124 .060 During the last week I was bored and I hated my life .654 .023 .196 During the last week I didn’t have any hope for the future .512 .164 .182 -.018 -.291 During the last week I was fighting for no particular reason During the last week I was feeling comfortable and secure wherever I went** .533 -.392 .086 .529 During the last week I was bored and I had nothing to do .531 -.213 During the last week I felt like hitting someone** .384 -.408 During the last week I was having thoughts of death .502 During the last week I was feeling emotionally drained .633 .354 .494 .261 .130 -.233 -.061 .049 -.044 .078 .093 .019 -.073 .106 .013 .128 -.024 -.011 .283 .082 -.041 .159 .423 .192 .074 .053 .279 -.160 .040 -.198 .073 -.366 .184 -.207 -.062 .267 .095 .218 .188 -.071 -.181 During the last week my heart was beating fast even without doing any type of sports .515 -.039 -.253 .199 .259 .128 .232 During the last week nothing mattered for me/I didn’t care for anything** .323 -.383 -.020 .120 -.039 .139 .575 During the last week I was happy** .257 .469 -.059 .027 .138 -.105 .116 .353 .030 -.128 -.105 .304 .214 -.282 .540 During the last week I was feeling fidgety, and moving a lot. I couldn’t sit still for a long .566 -.189 time without any particular reason (for example without having an exam) During the last week I was having a lot of headaches, stomachaches and nausea .637 -.160 .325 .040 Extraction method: principal component analysis a 7 components extracted ** excluded from final scale 123 338 Fig. 2 Results of the construct validity correlating self rated general health and self rated mental health with the AYMH scale 7—‘‘feeling secure,’’ ‘‘feeling happy,’’ and ‘‘not caring about anything’’—could not be theoretically/substantively described in a coherent theme. Therefore, all 7 items were excluded and only the 21 items that loaded well on factor 1 were kept in the scale. As such, factor analysis was an important data reduction step and served to condense the mental health scale to a linear combination of 21 items that loaded relatively well on one factor. The internal validity of the scale was high with a cronbach’s alpha of 0.901. As to the construct validity of the scale, results of ANOVA tests of unequal means clearly revealed that adolescents in the sample who rated their GSRH and SRMH (these two questions were added to the questionnaire) poorly scored low on the mental health scale. Figure 2 shows that the means were different from each other, and this difference was significant at the P \ 0.001 for both GSRH and SRMH. Furthermore, the figure shows a clear gradient in mean differences. In sum, the internal and construct validation tests revealed that the mental health scale of 21 items generated through an iterative process was valid in measuring mental health among Palestinian refugee adolescents. It can be used as an evaluation instrument in future research with this group. Discussion The authors engaged in the process described herein because of the scarcity of critical analysis and research on mental health scales that are culturally relevant and psychometrically robust in the Arab context generally and for Palestinian youth specifically. We recognize that there is not one single perfect scale, however, the objective was to find one which most closely matched the research 123 Community Ment Health J (2011) 47:331–340 objectives, all the while being structurally acceptable, appropriate, suitable, and applicable to the target group. A similar process of identifying a scale conducted by a different research group for different research objectives and a different target population could lead to the selection of a different scale. This critical analysis process, however, ensures that a more relevant and meaningful assessment instrument is selected. Researchers and practitioners working with youth should be sensitive to the particularities of the specific context and population they work within in order to avoid utilizing an assessment tool that is not pertinent to their purposes. The interactive process of translating, soliciting feedback from youth groups, the community coalition and professionals, and meticulous field testing proved to require much attention to detail, group work and decision making. However, the outcome was a contextually sensitive and culturally appropriate reliable scale to measure mental health of young Palestinian refugees. The research team members were familiar with the target group, the planned intervention, and the community context where the intervention would be conducted, all of which are necessary criteria for such a process to succeed. As a result of this process of validation, the Arab Youth Mental Health scale is grounded in community yet meets professional criteria for measuring common mental disorders. We believe that a final step in the validation of this scale should be to clinically validate it against a psychiatric interview using diagnostic criteria. Since the factor analysis described herein was exploratory, we also recommend further analysis to confirm the findings. There are several limitations to the work described in this manuscript. All the work leading to the development and validation of this scale was conducted in one refugee camp as part of a larger intervention research project. The youth who provided their understanding of scale items (in the focus groups) as well as descriptions of their feeling and actions when distressed were between the ages of 10 and 14 years. We engaged both out of school youth (13–19 years) and youth in schools (10–14 years) of both genders. The age groups were different, and the current AYMH scale is for use with the younger age group (10–14 year olds). The scale was not correlated against another mental health scale. We could not find any scales that were similar and had been tested for psychometric properties in Arabic. However, despite the fact that this scale was developed for use with Palestinian youth, we believe it has relevance and applicability to most Arab youth. Although the context of a refugee camp may be different than that of non-refugee communities or neighborhoods, many of the political and social factors influencing youth of the camp also influence Arab youth in general to a different degree. The scale also Community Ment Health J (2011) 47:331–340 uses very simple language and uses Arabic terms that are understandable; therefore it can be used with youth in or out of school. We suggest that researchers or practitioners working in other developing world contexts consider whether the AYMH scale or other scales are appropriate to the peculiarities of their cultural situations. If not, then we recommend that they engage in a similar process to develop and validate a mental health measure, using quantitative and qualitative data as well as community and professional feedback. Acknowledgments This paper was produced in the framework of a larger, inter-disciplinary research project on Urban Health, coordinated by the Center for Research on Population and Health at the Faculty of Health Sciences, American University of Beirut, Lebanon, with generous support from the Wellcome Trust, Mellon Foundation, and Ford Foundation. The authors would like to thank Dr. Trudy Harpham for her valuable guidance in reviewing earlier drafts of this paper, and Tanya Salem for her help with the development of the diagram. The authors would also like to thank the community and youth of Burj El Barajneh Palestinian refugee camp in Beirut as well as Drs. Hassen El Amin, Brigitte Khoury, Tima El Jamil, and Madeleine Badaro. References Abdel-Khalek, A. M. (2002). Why do we fear death? The construction and validation of the reasons for death fear scale. Death Studies, 26(8), 669–680. Afifi, R., Nakkash, R., & Haddad, P. (2010). Determinants of mental health among Palestinian youth. In Paper Presented at the UAEU Conference on ‘‘Global Health and the UAE’’, Al Ain, Abu Dhabi. Baker, A. M., & Kanan, H. M. (2003). Psychological impact of military violence on children as a function of distance from traumatic event: The Palestinian case. Intervention, 1(3), 13–21. Bean, T., Derluyn, I., Eurelings-Bontekoe, E., Broekaert, E., & Spinhoven, P. (2007). Validation of the multiple language versions of the Hopkins symptoms checklist-37 for refugee adolescents. Adolescence, 42(165), 51–71. Bowling, A. (2001). Measuring disease. Buckingham, Philadelphia: Open University Press. Bowling, A. (2005). Measuring health. New York: Open University Press, McGraw-Hill. Boyle, M. H., & Jones, S. C. (1985). Selecting measures of emotional and behavioral disorders of childhood for use in general populations. Journal of Child Psychology and Psychiatry, 26, 137–159. Bradburn, N. M. (1969). The structure of psychological well-being. Chicago: Aldine. Bryant, F. B., & Yarnold, P. R. (2000). Principal-components analysis and exploratory and confirmatory factor analysis. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding multivariate statistics (pp. 99–136). Washington, DC: American Psychological Association. Costello, E. J., & Angold, A. (1998). Scales to assess child and adolescent depression: Checklists, screens, and nets. Journal of the American Academy of Child and Adolescent Psychiatry, 27(6), 726–737. 339 Elbedour, S., Bart, W., & Hektner, J. (2007). The relationship between monogamous/polygamous family structure and the mental health of Bedouin Arab adolescents. Journal of Adolescence, 30(2), 213–230. Foldspang, A., & Montgomery, E. (2000). Criterion-validity-based assessment of four scale constructs. Scandinavian Journal of Public Health, 28(2), 146–153. Giacaman, R. (2004). Psycho-social/mental health care in the occupied Palestinian territories: The embryonic system. Birzeit: Birzeit University, Institute of Community and Public Health in cooperation with the Center for Continuing Education, Birzeit University. Giacaman, R., Saab, H., Nguen-Gillham, V., Abdullah, A., & Naser, G. (2004). Palestinian adolescents coping with trauma. The Occupied Palestinian Territory: Institute of community and Public Health. Birzeit University. Goldberg, D. P., & Hiller, V. F. (1979). A scaled version of the general health questionnaire. Psychological Medicine, 9, 139– 145. Harding, T. W., De Arango, N., & Baltazar, J. (1980). Mental disorders in primary health care: A study of their frequency and diagnosis in four developing countries. Psychological Medicine, 10, 231–241. Harpham, T., Reinchenheim, R. O., Thomas, E., Hamid, N., Jaswal, S., Ludermir, A., et al. (2003). Measuring mental health in a cost-effective manner. Health Policy and Planning, 18(3), 344– 349. Hundt, G. L., Chatty, D., Thabet, A. A., & Abuateya, H. (2004). Advocating multi-disciplinarity in studying complex emergencies: The limitations of a psychological approach to understanding how young people cope with prolonged conflict in Gaza. Journal of Biosocial Science, 36(4), 417–431. Jacobsen, L. B. (2000). Finding means: UNRWA’s financial situation and the living conditions of Palestinian refugees (Summary Report, Fafo-report 415). Norway: Interface Media. Kessler, R., & Mroczek, D. (1994). Final version of our non-specific psychological distress scale. Ann Arbor (MI), Survey Research Center of the Institute for Social Research: University of Michigan. Keyes, E. F. (2000). Mental health status in refugees: an integrative review of current research. Issues in Mental Health Nursing, 21, 397–410. Makhoul, J. (2003). Physical and social contexts of the three urban communities of Nabaa, Borj el Barajneh Palestinian Camp and Hay el Sullum (Unpublished report). CRPH: American University of Beirut. Makhoul, J., & Nakkash, R. (2009). Understanding youth: Using qualitative methods to verify quantitative community indicators. Health Promotion Practice, 10(1), 128–135. Mattsson, N. B., Williams, H. V., Rickels, K., Lipman, R. S., & Uhlenhuth, E. H. (1969). Dimensions of symptom distress in anxious neurotic outpatients. Psychopharmacology Bulletin, 5, 19–32. McIntyre, A. (2000). Constructing meaning about violence, school, and community: Participatory action research with urban youth. Urban Review, 32(2), 123–154. Mollica, R. F., Caspi-Yavin, Y., Bollini, P., Truong, T., Tor, S., & Lavelle, J. (1992). The Harvard trauma questionnaire: Validating a cross sectional instruments for measuring torture, trauma, and posttraumatic stress disorder in Indochinese refugees. Journal of Nervous and Mental Disease, 180, 111–116. Mollica, R. F., Poole, C., Son, L., Murray, C. C., & Tor, S. (1997). Effects of war trauma on Cambodian refugee adolescents’ functional health and mental health status. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 1098–1106. 123 340 Moreau, N., Hassan, G., Rousseau, C., & Chenguiti, K. (2009). Perception that ‘‘everything requires a lot of effort’’: Transcultural SCL-25 item validation. Journal of Nervous and Mental Disease, 197(9), 695–699. Myers, K., & Winters, N. C. (2002a). Ten-years of rating scales I: Overview of scale functioning, psychometric properties, and selection. Journal of American Academy for Child and Adolescent Psychiatry, 41(2), 114–122. Myers, K., & Winters, N. C. (2002b). Ten-year review of rating scales II: Scales for internalizing disorders. Journal of American Academy for Child and Adolescent Psychiatry, 41(6), 634–659. Nese, E., Simsek, Z., Oner, O., & Munir, K. (2005). Effects of internal displacement and resettlement on the mental health of Turkish adolescents’. European Psychiatry, 20(2), 152–157. O’Donnell, J., Michalak, E. A., & Amer, E. B. (1997). Inner-city youths helping children: After-school programs to promote bonding and reduce risk. Social Work in Education, 19(4), 231– 242. Ommeren, M. V. (2003). Validity issues in trans-cultural epidemiology. British Journal of Psychiatry, 182, 376–378. Parkerson, G. R., Broafhead, W. E., & Tse, C.-K. J. (1990). The duke health profile: A 17-item measure of health and dysfunction. Medical Care, 28(11), 1056–1072. Patel, V., Flisher, A. J., Hetrick, S., & McGorry, P. (2007). Mental health of young people: A global public health challenge. Lancet, 369, 1302–1313. Punamaki, R. L., Komproe, I. V., Qouta, S., Elmasri, M., & De Jong, J. (2005). The role of peritraumatic dissociation and gender in the association between trauma and mental health in a Palestinian community sample. American Journal of Psychiatry, 162, 545–551. Radloff, L. S. (1977). The CES-D scale: A new self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. Rousseau, C., & Drapeau, A. (1998). Parent-child agreement on refugee children’s psychiatric symptoms: A transcultural perspective. Journal of the American Academy of Child and Adolescent Psychiatry, 37(6), 625. Savin, D., Sack, W. H., Clarke, G. N., Meas, N., & Richart, I. (1996). The Khmer Adolescent Project III: A study of trauma from Thailand’s site II refugee camp. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 384–391. Schopler, E., Reichler, R. J., DeVellis, R. F., & Daly, K. (1980). Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). Journal of Autism and Development Disorders, 10(1), 91–103. Slodniak, V. (2002). Psychosocial functioning of Bosnian refugee adolescents in Slovenia. In S. Powell & E. Durakovic-Belko (Eds), The psychosocial consequences of war: Results of empirical research from the territory of former Yugoslavia (pp. 198–200). UNICEF: Sarajevo, Bosnia-Herzegovina. Spector, P. E., & Fox, S. (2003). Reducing subjectivity in the assessment of the job environment: Development of the Factual Autonomy Scale (FAS). Journal of Organizational Behavior, 24, 417–432. Spector, P. E., & Jex, S. M. (1998). Development of four self-report measures of job stressors and strain: Interpersonal conflict at 123 Community Ment Health J (2011) 47:331–340 work scale, organizational constraints scale, quantitative workload inventory, and physical symptoms inventory. Journal of Occupational Health Psychology, 3(4), 356–367. Tennant, R., Hiller, L., Fishwick, R., Platt, S., Joseph, S., Weich, S., Parkinson, J., Secker, J., & Stewart-Brown, S. (2007). The Warwick-Edinburgh Mental Well-being Scale (WEMWBS): Development and UK validation. Health and Quality of Life Outcomes, 5, 63–75. doi:10.1186/1477-7525-5-63. Thabet, A. M., Abed, Y., & Vostanis, P. (2002). Emotional problems in Palestinian children living in a war zone: A cross sectional study. Lancet, 359, 1801–1804. Thabet, A. A., Stretch, D., & Vostanis, P. (2000). Child mental health problems in Arab children: Application of the strengths and difficulties questionnaire. International Journal of Social Psychiatry, 46(4), 266–280. Thabet, A. A., & Vostanis, P. (1998). Social adversities and anxiety disorders in the Gaza Strip. Archives of Disease in Childhood, 78(5), 439–442. Thabet, A. A., & Vostanis, P. (1999). Post traumatic stress reactions in children of war. Journal of Child Psychology and Psychiatry, 40(3), 385–391. Thabet, A. A., & Vostanis, P. (2001). Epidemiology of child mental health problems in Gaza Strip. Eastern Mediterranean Health Journal, 7(3), 403–412. UNRWA. (2009). UNRWA-Lebanon, facts & figures 2009. Retrieved from http://www.unrwa.org/etemplate.php?id=65. Accessed April 9, 2010. Veit, C. T., & Ware, J. E. (1983). The structure of psychological distress and well-being in general populations. Journal of Consulting and Clinical Psychology, 51, 730–742. Viswanathan, M., Ammerman, A., Eng, E., Gartlehner, G., Lohr, K.N., Griffith, D., Rhodes, S., Samuel-Hodge, C., Maty, S., Lux, L., Webb, L., Sutton, S.F., Swinson, T., Jackman, A., & Whitener, L. (2004). Community-based participatory research: Assessing the evidence (Evidence Report/Technology Assessment No. 99). RTI: University of North Carolina, Evidencebased Practice Center under Contract No. 290-02-0016. AHRQ Publication 04-E022-2. Rockville, MD: Agency for Healthcare Research and Quality. Vostanis, P. (2006). Strengths and difficulties questionnaire: Research and clinical applications. Current Opinion in Psychiatry, 19, 367–372. Ware, J. E., & Sherbourne, C. D. (1992). The MOS 36-item shortform health survey (SF-36): I. Conceptual framework and item selection. Medical Care, 30, 473–483. Ware, J. E., Sherbourne, C. D., & Davies, A. R. (1992). Developing and testing the MOS 20-item short-form health survey: A general population application. In A. L. Stewart & J. E. Ware (Eds.), Measuring functioning and well-being: The medical outcomes study approach (pp. 277–290). Durham: Duke University Press. Winters, N. C., Myers, K., & Proud, L. (2002). Ten-year review of rating scales. III: Scales assessing suicidality, cognitive style, and self esteem. Journal of the American Academy of Child and Adolescent Psychiatry, 41(10), 1150–1181. World Health Organization Quality Of Life Group. (1998). Development of the world health organization WHOQOL—bref quality of life assessment. Psychological Medicine, 28, 551–558.