Screening for Psychopathology Using theThree Factors Model of the Structure of Psychopathology: A Modified Form ofGAIN Short Screener

The goal of this paper is to develop a valid and reliable screening tool for mental health that is based on empirically and conceptually valid structure of psychopathology. Recently several studies of the structure of psychopathology found a general factor and three specific factors: internalizing, externalizing and thought disorder. We adapted the previously validated GAIN Short Screener to include the thought disorder that was not included in its original version and further developed its internalizing subscale. We conducted an exploratory and confirmatory factor analysis of the new adapted measure and produced 20 items screening tool that parsimoniously represents the three factors. The adapted screener and its subscales were found to have good reliability, stability, structural validity in two Egyptian and Polish samples. Additionally, all its subscales significantly correlated with different trauma types and with cumulative trauma, and negatively with self-esteem. The new adapted measure is the first that is based on robust scientific evidence of the structure of psychopathology and can be used in a broad scope of settings.


Introduction
There is a lack of valid and reliable screening tool for psychopathology that is How to cite this paper: Kira, I. A., Shuwiekh, H., & Kucharska, J. (2017).Screening for Psychopathology Using the Three Factors Model of the Structure of Psychopathology: A Modified Form of GAIN Short Screener. Psychology, 8, 2410-2427.https://doi.org/10.4236/psych.2017.814152based on robust conceptual and empirical evidence of the structure of psychopathology.There were no empirically validated conceptual models behind most of the existing measures.Most screening measures for psychopathy targeted either specific disorder or general psychopathology.Most of the measures that screen for general psychopathy either utilized the diagnostic criteria of mental disorder (e.g., Harvard trauma questionnaire) or targeted the general psychopathology in aparticular population (e.g., refugees) (e.g., cumulative trauma disorders in refugees, Kira et al., 2012).World Health Organization WHO (Beusenberg, Orley, & World Health Organization, 1994) developed a self-reporting questionnaire of 20 questions (SRQ-20) as a screening tool to detect common mental disorders (CMD).Several versions of the Self-Reporting Questionnaire (SRQ) were used in screening and research.SRQ is not based on empirical or theoretical analysis of the structure of psychopathology.It includes only symptoms related to anxiety and depression.The mood, neurotic and psychotic disorders are also common and there is a noticeable overlap of symptoms of depression, anxiety, fatigue, or somatic complaints in CMD.However, different versions added other items that represented psychotic symptoms (e.g., Youngmann et al., 2008).One of the measures that widely used with refugees and torture survivors is Harvard trauma questionnaire (HTQ) (e.g., Mollica et al., 1992).HTQ may be a useful tool for measuring some syndromes, but not designed to be a comprehensive screening tool for psychopathology.The same critique that targeted early SRQ versions applies to HTQ, as it does not measure, for example, dissociation psychosis and other mental health syndromes especially present in multiply traumatized populations (Kira et al., 2012).
Co-morbidity of mental disorders is commonly found in clinical and epidemiological studies (e.g., Kessler, Chiu, Demler, & Walters, 2005;Angold, Costello, & Erkanli, 1999).Research suggests the existence of a general psychopathology factor, which is associated with high risk of developing a broad range of internalizing, externalizing and psychotic mental disorders (e.g., Lahey et al., 2012).In one study, a general latent factor based on repeated assessments of psychiatric symptoms over a 20-year period explained on average 42% of the disorders variance (Caspi et al., 2014;Carragher, Krueger, Eaton, & Slade, 2015).
In another large multi-ethnic adult sample, a general factor was estimated to explain between 29% and 67%, depending on the diagnosis (Kim & Eaton, 2015).
The general psychopathology factor was associated with lower IQ, higher negative affectivity, and lower effortful control (Neumann et al., 2016).Importantly, the general psychopathology factor showed a significant Single nucleotide polymorphism (SNP) heritability of 38% (Neumann et al., 2016).Most of the studies above used DSM oriented scales; however, the general psychopathology factor was also replicated in studies using problem scales/items in general population samples (Laceulle, Vollebergh, & Ormel, 2015;Murray, Eisner, & Ribeaud, 2016).These new advances in discovering the component of psychopathology structure gave us an opportunity to develop psychopathology screening tool for Psychology adult and adolescent from various populations, based on the robust empirically validated conceptual model of psychopathology that represent its three main factors: Internalizing, externalizing and thought disorder.To develop such a measure that represents the three factors, we previously adapted and utilized GAIN-Short Screeners (GAIN-SS) (Dennis, Chan, & Funk, 2006) in several studies.GAIN-SS was developed initially to screen for psychopathy in adults and adolescent and includes measures for externalizing, internalizing and addiction, but does not add a measure of thought disorder.GAIN-SS is a screener that identifies clients (adults and adolescents) who are likely to have mental health disorders, issues with crime/violence, and issues with substance use.In the first adaptation of the measure, we added items to the internalizing section that are related to posttraumatic stress disorder symptoms.The original version of internalizing subscale did not include different PTSD symptoms.We added a subscale for psychoticism and dissociation using items from psychoticism/dissociation subscale of cumulative trauma disorder scale (Kira et al., 2012).
The measure in its initially adapted from included the three primary components of psychopathology: Internalizing, externalizing and thought disorder (psychoticism) (e.g., Caspi et al., 2014;Laceulle, Volleberge, & Ormel, 2015).The initially adapted measure included 32 items (Internalizing: nine questions, Externalizing and Substance Abuse: 14 items, thought disorder or psychoticism: nine items) (see Appendix 1).The participant is asked to indicate if the behavior (or feeling) happened in the past month (scored 4), or occurred in the last 2 -3 months (scored 3), or in the last 3 -12 months (scored 2), or the last year or more (scored 1), or never happened (scored 0).High scores indicate potentially higher symptoms in these areas.
The authors utilized the version that has been previously adapted in several studies (e.g., Kira, Shuwiekh, & Bujold-Bugeaud, 2017;Kira, Shuwiekh, Kucharska, Abu-Ras, & Bujold-Bugeaud, 2017;Kucharska, 2017) and proved to be useful, reliable and valid.The goal was to further develop, refine and evaluate its psychometric properties.Accordingly, we are assessing this previously adapted screener that measure the three factors identified as the specific components of psychopathology to make it a more parsimonious and focused screening tool.
The goal is to trim the previously adapted version of the GAIN Short screener.
We initially deleted two items from the externalizing subscale to make it more parsimonious.We conducted our current analysis of the left 30 items.
Research Questions 1) Does the adapted version of GAINS screener have adequate reliability and stability?
2) Does it have good construct heuristics being associated with different trauma types and with cumulative trauma, and negatively associated with self-esteem?
3) Does exploratory and confirmatory factor analysis support the structural validity of the three specific factors of internalizing, externalizing and thought disorder in two Western (Polish), and non-Western (Egyptian) samples?

The First Sample (The Egyptian Sample)
1) Participants  the same measures that we will describe in the following section.

The Measures Used in the Two Studies
In addition to the modified version of the GAINS screener, the used measures included the following measures: The Cumulative Trauma Scale CTS-S (short form) is a measure based on the development-based trauma framework (DBTF) (e.g., Kira, 2001;Kira, Ashby et al., 2013;Kira, Fawzi, & Fawzi, 2013;Kira, Lewandowski et al., 2008;Kira, Lewandoski, Chiodo, & Ibrahim, 2014;Kira, Omidy, & Ashby, 2014).DBTF identifies and measures different dimensions of individual development that may be affected by stress and traumatic stress (i.e., attachment, personal, collective and role identities, and interdependence).The CTS-S is a 32-item instrument that measures cumulative trauma regarding the occurrence, frequency, type, and participant denotes that she/he has experienced the traumatic event, then he/she is asked to describe her/his appraisal of its effect on a 7-point Likert-type scale (1 = extremely positive; 7 = extremely negative).CTS-S includes two general subscales for cumulative trauma dose: occurrence and frequency of experience, and Psychology two appraisal subscales: negative and positive appraisal.Four subscales for each of the trauma types can be obtained.
The alpha for the scale of cumulative trauma occurrence was .88 in the Egyptian data and .91 in the Polish data.The measure was used to test if the adapted GAINShort Screener and its sub-tests will be significantly associated with different trauma types.
The Rosenberg self-esteem scale (RSES) is a 10-item scale that measures global self-esteem (Rosenberg, 2015).Each item rated on a 4-point Likert-type scale from strongly agree to disagree strongly and scored from 0 to 3. The scale divided into five positively worded and five negatively worded statements.The RSES has been translated and adapted to various languages including Arabic.
Rosenberg reported good psychometrics for the scale and its reliability ranging from .85 to .88.In previous Arabic samples, alpha was .75.Test-retest using an independent sample of 35 males with four weeks interval yielded excellent stability coefficient of .983.In the Egyptian study, its alpha was .72,In the Polish study, its alpha reliability was .78.The measure was used to test if the adapted GAINShort Screener and its sub-tests will be negatively associated with self-esteem.

Translation into Polish Procedures
Self-esteem scale: the Polish adaptation, was published in 2008 (Dzwonkowska, Lachowicz-Tabaczek, & Laguna, 2008), the scale has good reliability and validity and is widely used in Poland.For the other scales: first certified Polish transla-Psychology tors translated the tools into Polish, then were back-translated into English, a third expert compared the initial and final English versions.No significant differences found in the case of discrimination scales and authoritarianism scale.
Minor differences found in the cumulative trauma scale and Gain externalizing scale, but the third expert decided that the items have the same meaning as the words used have a similar semantic field.

Translation into Arabic Procedures
Some of these measures have been previously translated into Arabic and proved to have adequate reliability and validity in Arabic clients in previous studies, as will be briefly described when introducing them in the measures section.We translated the other measures (modified GAIN, and F scale) into Arabic.The committee that translated the measures consisted of three bilingual professionals who conducted the forward translation and two different bilingual professionals who contributed to the reverse translation.The translations compared, and the differences discussed until a consensus reached on the final version by the committee.

Statistical Analysis Strategy
The data were analyzed utilizing IBM-SPSS 22 and Amos 22 software.We split the Egyptian sample into two sub-samples (N = 261 each).We conducted exploratory factor analysis (Principal axis factoring method) of the Adapted GAINShort screening items in the first Egyptian sub-sample.We conducted exploratory (on the first sub-sample) and confirmatory factor analysis (on the second sub-sample).Because internalizing, externalizing and thought disorder are assumed to be correlated with a higher second-order factor, we conducted an oblique rotation.We used the scree test (Cattell, 1966) and parallel analysis (O'Connor, 2000) to help determine the number of factors.Confirmatory factor analysis was conducted on the resulted in three factors.Following Byrne's (2012) recommendations, the criteria for good model fit were a non-significant (χ2), (χ2/d.f.> 2), comparative fit index (CFI) values > 0.90, and root-mean-square error of approximation (RMSEA) values < 0.06.We investigated the reliability of the sub-scales with the Cronbach's alpha.To test its predictive validity, we conducted a zero-order correlation to explore the linear relationships between the measured constructs.

Reliability
In

Correlations
Externalizing, Internalizing, and thought disorder subscales correlated significantly with all trauma types and with cumulative traumas.All the three subscales correlated negatively with self-esteem.The three subscales were highly correlated.Table 1 provides the zero-order correlations between the mentioned variables in the Egyptian sample.Similar correlations between the three subscales and cumulative trauma, different trauma types were found in the Polish sample.
The correlation results provide initial evidence of predictive validity and the Construct Heuristics of the subscales.

Structural Validity
Exploratory factor analysis of both the Egyptian sub-sample and the Polish sample yielded three factors with all items loaded significantly on the first factor (before rotation) which may validate the one-factor solution of psychopathology obtained in previous studies.The Oblimin rotation produced three clear-cut factors that represented the three constructs (thought disorder, internalizing and externalizing (Table 2, see also Appendix 1).The three factors accounted for 48.29% of the variance.The first factor loaded on thought disorder items and accounted for 33.67% of the variance.The second factor loaded on internalizing items and accounted for 10.06% of the variance.The third factor loaded on externalizing items and suicidality and accounted for 4.56% of the variance.We deleted the items that have less than .40loadings as well as those that have cross-loading (10 items), and reanalyzed the remaining 20 items.The results indicated a clean three factors solution with a different order (Table 3, see also  Note: Bolded items are either have loadings less than .4 on the factor, or have cross-loading and were deleted in the second-factor analysis.Appendix 2).The analysis yielded three factors accounted for 54.64% of the variance.In this analysis the first factor included the items related to externalizing and accounted for 35.89% of the variance.The second factor included the items related to internalizing and accounted for 12.22% of the variance.The third factor included the items related to thought disorder and accounted for 6.54% of the variance.
Additionally, we conducted confirmatory factor analysis on the second Egyptian sample and the Polish sample.For the Egyptian sample, the three factors, twenty items structure fitted adequately (Chi Square = 322.117,d.f.= 159, p = .000.CFI = .932,RMSEA = .063). Figure 1 includes the results of the confirmatory factor analysis for GAIN-20 Short Screener in The Egyptian sample.The confirmatory factor analysis using the Polish sample did not fit adequately in the initial analysis.However, the modification indices strongly suggested moving the suicidality item from externalizing items to internalizing items.

Conclusion and Future Directions
We conclude that the modified GAIN-20 is a structurally valid tool for screening of mental health based on the rigor scientific evidence of the structure of psychopathology.The measure subscales have good reliability and stability in both Western and non-Western populations.The goal of this study was to adapt the GAIN Short Screener to measure the three factors that were found in the scientific studies of the structure of psychopathology.The objective was to establish a parsimonious screening tool that may be used to screen for psychopathology in adults and adolescents.We emphasized that the significance of current study lies in the fact that it is the first that provided a tool for mental health screening that is based on the scientific evidence of the structure of psychopathology

Limitations
While the current study was an important first step, it has several limitations.
For example, the fact that different traumatic stressors significantly correlated with the three subscales does not provide specific predictive validity information of the subscales.To establish their predictive validity, clinical samples should be used.Our samples were college students and it was difficult to establish the predictive validity of the sub-scales using non-clinical samples.Additionally, the externalizing subscales can be reconstructed to include more diverse items to be tested in future studies.The current study is an initial step in devising more accurate screening tool based on the scientific evidence of the structure of psychopathology.An expanded well-funded study may be needed to develop it further to increase the representativeness of its items of all aspects of psychopathology and to establish it as the standard screening measure for psychopathology in the field.Regardless, the modified GAIN-20 screener, in its current form is a valid and reliable tool to screen for psychopathology based on the rigor scientific evidence of its structure.

Appendix 1
Adapted GAIN Short Screener (A-GAIN-SS-30) for Internalizing, Externalizing, and Thought Disorder (The first modified version).
The following questions are about common psychological, behavioral, and personal problems.These problems are considered significant when you have them for two or more weeks when they keep coming back, when they stop you from meeting your responsibilities, or when they make you feel like you can't go on.
After each of the following questions, please tell us the last time, if ever, you had the problem by answering whether • It was in the past week (5) • It was in the past month (4), • 2 to 3 months ago (3), • 4 to 12 months ago (2), • 1 or more years ago (1), • Never (0).
1.When was the last time that you had significant problems with: a) Feeling very trapped, lonely, sad, blue, depressed, or hopeless about the future?4 3 2 1 0 b) Sleep trouble, such as bad dreams, sleeping restlessly, or falling asleep during the day? 4 3 2 1 0 c) Feeling very anxious, nervous, tense, scared, panicked, or like something bad was going to happen? 4 3 2 1 0 d) Becoming very distressed and upset when something reminded you of the past? 4 3 2 1 0 e) Thinking about hurting self, ending your life or committing suicide? 4 3 2 1 0 f) Having unexpected or disturbing memories?4 3 2 1 0 g) Trying to avoid reminders of painful past events?4 3 2 1 0 h) Jumping or being very frightened by sudden loud noises? 4 3 2 1 0 i) Feeling out of touch with surrounding? 4 3 2 1 0 2. When was the last time that you did the following things two or more times?a) Lied or conned to get things you wanted or to avoid having to do something 4 3 2 1 0 b) Had a hard time listening to instructions at school, work, or home 4 3 2 1 0 c) Had a hard time waiting for your turn 4 3 2 1 0 d) Bullied or threatened other people 4 3 2 1 0 e) Started physical fights with other people 4 3 2 1 0 f) Took something from a store without paying for it?4 3 2 1 0 g) Lost temper 4 3 2 1 0 h) Easily irritated 4 3 2 1 0 i) Failed to respect those who may represent authority 4 3 2 1 0 j) Get irritated, to the extent you do not care about safety 4 3 2 1 0 3.When was the last time that: a) you spent a lot of time either getting alcohol or other drugs, using alcohol or other drugs or recovering from the effects of alcohol or other drugs (e.g., feeling sick)? 4 3 2 1 0 Psychology 5 4 3 2 1 0 5) Were a bully or threatened other people.5 4 3 2 1 0 6) Thinking about hurting self, ending your life or committing suicide 5 4 3 2 1 0 7) Took something from a store without paying for it?5 4 3 2 1 0 B. When was the last time that you had significant problems with?8) Feeling very anxious, nervous, tense, scared, panicked, or like something bad was going to happen 5 4 3 2 1 0 9) Sleep trouble, such as bad dreams, sleeping restlessly, or falling asleep during the day 5 4 3 2 1 0 10) Becoming very distressed and upset when something reminded you of the past 5 4 3 2 1 0 11) Trying to avoid reminders of painful past events 5 4 3 2 1 0 12) Feeling very trapped, lonely, sad, blue, depressed, or hopeless about the future 5 4 3 2 1 0 13) Having unexpected or disturbing memories 5 4 3 2 1 0 C. When was the last time that you? 14) Felt as if you are almost two different people? 5 4 3 2 1 0 15) Felt that you do not have enough control over your responses and reactions 5 4 3 2 1 0 16) Felt apathetic, with no emotion 5 4 3 2 1 0 17) Felt people/enemies are following you any place you go 5 4 3 2 1 0 18) Felt you are in two or more different places in the same time 5 4 3 2 1 0 19) Get irritated, to the extent you do not care about safety 5 4 3 2 1 0 20) Seeing or hearing things that no one else could see or hear 5 4 3 2 1 0 the questionnaire to participants in Arabic from September to November of 2015.While the questionnaires administered to males and females were different, all have the same measures utilized in the current analysis and were combined in one data set.The participation was voluntary.Each participant was informed about the general goals of the study and signed informed consent to participate.Each person took between 40 -50 minutes to complete the questionnaire.The Institution Review Board of authors' institution approved the research as part of a cross-cultural study of gender relations among college students.included combined two data sets (one included males and the other females only) of 467 college students from two Polish cities: Warsaw and Wrocław.Females were 59.3% of the sample.Study assistants recruited participants via opportunity sampling at university campuses during the breaks between classes.The age range in the sample is 18 -34, M = 22.39,SD = 2.81.All participants were residents of one of the cities and enrolled as students at the Psychology time of the data collection.Participants signed an informed consent and received no compensation for their participation in the study.2) Procedure Research team administered the questionnaire in Polish, from January to February 2016.Each participant informed about the general goals of the study and signed informed consent to participate.Each person took between 40 -50 minutes to complete the questionnaire.While the questionnaires administered to males and females were different, all have the same measures utilized in the current analysis and were combined in one data set.The two data sets included negative and positive appraisals.The test is intended to measure at least seven major trauma types.They include collective identity trauma (3 items), personal identity trauma (6 items), survival trauma (6 items), attachment trauma (2 items), secondary trauma (7 items), achievement traumas (2 items) and gender discrimination (2 items).Collective identity trauma includes trauma related to exposure to war and torture and discrimination based on race, ethnicity, or national origin.Personal identity trauma includes trauma related to sexual abuse, rape, incest, and being robbed.Attachment trauma comprises abandonment by parents.Survival trauma includes car accidents, life-threatening illnesses, and natural disasters.Achievement or role identity trauma is intended to measure traumatic stressors related to attainment of life goals like success in school or business.Secondary trauma includes trauma related to having witnessed a traumatic event occurring to another individual or group and affecting social interdependence.Gender discrimination includes gender discrimination by parents (family) and gender discrimination by society and institutions.Gender discrimination items are worded to apply to both genders.In response to each item on the measure, participants are instructed to indicate their experience with a traumatic event on a 5-point Likert-type scale (0 = never; 4 = many times).If a

Figure 1 .
Figure 1.Confirmatory factor analysis for GAIN-20 short screener in the Egyptian sample.

Table 1 .
Zero order correlations between the modified GAINS' subscales, self-esteem and different trauma types in the Egyptian sample.

Table 2 .
Factor structure of the modified GAIN-30 screener in the Egyptian sub-sample.

Table 3 .
Factor structure of the modified GAIN-20 screener in the Egyptian sub-sample.