Psychometric Properties of the Icelandic Version of the ...³sa... · spurningalistum, BAI, CORE-OM og QOLS, auk þess að svara GAD-7 í annað sinn. Innri áreiðanleiki var talinn

Psychometric Properties of the Icelandic Version of the

Generalized Anxiety Disorder-7

Rósa Ingólfsdóttir

2014 BSc in Psychology

Author: Rósa Ingólfsdóttir ID number: 010688-2059 Department of Psychology School of Business

PSYCHOMETRIC PROPERTIES OF THE ICELANDIC GAD-7 2

Foreword

Submitted in partial fulfillment of the requirements of the BSc Psychology degree,

Reykjavík University, this thesis is presented in the style of an article for submission to a

peer-reviewed journal.


Abstract

The Generalized Anxiety Disorder-7 (GAD-7) is a self-report anxiety scale that screens for

Generalized Anxiety Disorder. The aim of this study was to assess the psychometric

properties of the Icelandic version of the GAD-7. The study included 358 participants

consisting of two samples, one clinical and one non-clinical. All participants completed the

GAD-7 and PHQ-9 questionnaires and participants in the clinical sample answered diagnostic

specific questionnaire based on main diagnosis evaluation as well. The participants in the

non-clinical sample more over completed additional questionnaires, BAI, CORE-OM and

QOLS, as well as GAD-7 on a second occasion. Internal reliability was good and both

convergent validity and divergent validity were supported. The comparison of participants in

the clinical sample and participants in the non-clinical sample showed that there was a

statistically significant difference between the two groups on scores on the questionnaires.

Patients diagnosed with anxiety disorder more over had higher scores on GAD-7 than

patients without such diagnosis, indicating good discriminant validity of the GAD-7. It is

concluded that the psychometric properties of the Icelandic version of the GAD-7 are

satisfactory.

Útdráttur

GAD-7 er sjálfsmatskvarði sem skimar fyrir einkennum almennrar kvíðaröskunar. Markmið

þessarar rannsóknar var að kanna próffræðilega eiginleika íslenskrar útgáfu GAD-7.

Þátttakendur í rannsókninni voru 358 einstaklingar í tveimur hópum, tilraunahóp og

samanburðarhóp. Allir þátttakendur svöruðu GAD-7 og PHQ-9 spurningalistunum og auk

þess svöruðu þátttakendur í tilraunahópnum sérstökum spurningalistum byggðum á

greiningarmati. Þátttakendur í samanburðarhóp svöruðu jafnframt þremur öðrum

spurningalistum, BAI, CORE-OM og QOLS, auk þess að svara GAD-7 í annað sinn. Innri

áreiðanleiki var talinn góður og niðurstöðurnar studdu bæði samleitniréttmæti og

aðgreiningarréttmæti GAD-7. Samanburður á þátttakendum í hópunum tveimur sýndi að það

var munur á milli stigaskors þeirra á spurningalistunum. Jafnframt voru sjúklingar með

kvíðagreiningu með hærra skor á GAD-7 en sjúklingar án kvíðagreiningar, sem bendir til

góðs aðgreiningarréttmætis GAD-7. Íslenska útgáfan af GAD-7 er talin hafa viðunandi

próffræðilega eiginleika.


Introduction

Anxiety disorders are an umbrella term that include a variety of disorders that share

common features of excessive fear and anxiety and related behavioral disturbances

(American Psychiatric Association, 2013). The diagnostic and statistical manual of mental

disorders fifth edition (DSM-5) identifies seven main types of anxiety disorders and one of

the most common is Generalized Anxiety Disorder (GAD) (American Psychiatric

Association, 2013; Spitzer, Kroenke, Williams, & Löwe, 2006). GAD involves chronic

worrying, nervousness and tension and DSM-5 states that individuals have to experience

excessive anxiety and worry, that they find difficult to control, for at least six months to be

diagnosed (American Psychiatric Association, 2013; Hansell, 2005). Given the estimated

prevalence of GAD, that is 2.8% to 8.5% in general practice and 0.4% to 5.7% among the

general population, it is important to have measuring devices that can assess GAD in an

accessible and reliable way (American Psychiatric Association, 2013; Butcher, Mineka, &

Hooley, 2013; Hansell, 2005; Roy-Byrne & Wagner, 2004). One such measuring device is

the Generalized Anxiety Disorder-7 (GAD-7), a 7-item self-report anxiety scale that screens

for GAD (Spitzer et al., 2006). The GAD-7 does not only screen for GAD but has as well

proved to be a good screening device for probable cases of panic disorder, social anxiety

disorder and post-traumatic stress disorder (PTSD) (Kroenke, Spitzer, Williams, Monahan, &

Lowe, 2007). The fact that the scale does not require a clinician administration makes it more

desirable since other diagnostic interviews, such as Primary Care Evaluation of Mental

Disorders (PRIME-MD), The Structured Clinical Interview (SCID) and The Mini-

International Neuropsychiatric Interview (MINI), are more time consuming and that alone

can create problems in primary care settings where most visits are 15 minutes or less

(Delgadillo et al., 2012; Sheehan et al., 1998; Spitzer et al., 1994; Spitzer, Kroenke, &


Williams, 1999; Spitzer RL, Williams JW, Gibbon M, & First MB, 1992). Recent research

has showed that the GAD-7 can be self-administered in less than 5 minutes.

What is demanded within the field is reliable and valid measures for anxiety disorders in

clinical strategy, especially when there is high prevalence and disability (Delgadillo et al.,

2012). The psychometric properties of the GAD-7 have been assessed and when conducting

the list Spitzer et al. assessed the reliability and validity of the measuring device (2006). The

results showed that the internal consistency was excellent (Cronbach α = 0.92) and the test-

retest reliability was good (intraclass correlation = 0.83). The convergent validity of the

GAD-7 was furthermore proved by its correlations to Beck Anxiety Inventory (BAI) (r =

0.72) and the anxiety subscale of the Symptom Checklist 90 (r = 0.74). The scale was

furthermore strongly correlated with the PHQ-8 depression measure (r = 0.75), which

includes all items from the PHQ-9 except for the item about suicidal ideation, and that is

consistent with results from previous studies of anxiety and depression. Previous research has

stated that the psychometric properties of the GAD-7 are good and the scale is clinically

useful and therefore a good measuring device for GAD (Delgadillo et al., 2012; Löwe et al.,

2008). The GAD-7 has been translated into several languages, e.g. Dutch, Spanish, German

and Turkish, and all items on the translated scales measure the same concepts as the original

one (Donker, van Straten, Marks, & Cuijpers, 2011; Garcia-Campayo et al., 2010; Konkan,

Senormanci, Guclu, Aydin, & Sungur, 2013; Löwe et al., 2008). The adapted versions all

showed excellent psychometric properties and the validity and reliability of the translated

versions of the GAD-7 have been confirmed. The GAD-7 has been translated into Icelandic

and in the present study the psychometric properties of the Icelandic version were assessed

for the first time.


Method

Participants

The study included 358 participants consisting of two samples, one clinical sample and

one non-clinical sample. The clinical sample consisted of 233 individuals, 192 females

(82.4%) and 41 males (17.6%), the mean age was 37.4 years (SD: 13.29, range 18 – 77

years). All participants were experiencing mental health problems as evaluated by a General

Practitioners (GPs) or a clinical psychologist administering MINI and all attended a

transdiagnostic cognitive behavioral group therapy (TCBGT) that was offered in 9 primary

health care centers in Reykjavík and surrounding towns. All referrals came from GPs and

criteria for referral was being over 18 years of age and showing signs of emotional problems

based on the GP clinical evaluation. Exclusion criteria for the treatment were: 1) obvious

signs of dementia or another generalized cognitive impairment, 2) presence of symptoms

suggesting current psychotic condition and 3) current self-reported substance abuse.

Non-clinical sample consisted of 125 individuals, 94 sports science students, 57 females

and 37 males, at Reykjavík University and 31 employees, 17 females and 14 males, at a

leisure center in Reykjavík. The mean age for the students’ sample was 25 years (SD = 3.4,

range 20 – 39 years) and for the employees it was 27 years (SD = 8.0, range 20 – 64 years).

The study used a convenience sample and there was no exclusion criteria.

Of the total sample, 264 (74%) participants were included in the study. The return of

completed questionnaires among the non-clinical sample was 125 (100%) and in the clinical

sample 139 (60%) patients fulfilled the study’s inclusion criteria and attended at least one

treatment session. Of the 125 individuals in the non-clinical sample 73 (58%) returned

completed questionnaire two weeks later for test-retest analysis.


Measures

GAD-7 is a self-report scale that is used to screen for and measure the severity of GAD

(Spitzer et al., 2006). Even though the scale was developed to screen for GAD it has proven

to be a good screener for probable cases of PTSD, social anxiety disorder and panic disorder.

The GAD-7 tests seven items and asks respondents how often, over the last two weeks, they

have been bothered by certain problems, e.g. not being able to sleep or control worrying or

feeling nervous, anxious or on edge (Appendix I). Response options are “not at all”, “several

days”, “more than half the days” and “nearly every day” and the answers are scored as 0, 1,

2, and 3. The total score of the scale ranges from 0 to 21 and cut points might be interpreted

as 5 being mild, 10 being moderate and 15 being severe levels of anxiety. Research has

showed that the internal consistency of the GAD-7 is excellent (Cronbach α = 0.92) and the

test-retest reliability is good (intraclass correlation = 0.83) (Spitzer et al., 2006). The results

from Spitzer’s et al. study revealed a strong association between increasing GAD-7 severity

scores and declining functional status, which indicates good construct validity. The results

from the same study furthermore showed that the convergent validity of the scale was good.

This was shown by its correlations to BAI (r = 0.72) and the anxiety subscale of the

Symptom Checklist 90 (r = 0.74). The divergent validity was also good where more than half

of patients with high anxiety scores did not have high depression scores.

Patient Health Questionnaire-9 (PHQ-9) is used to screen and assess depression severity

(Kroenke & Spitzer, 2002). The scale has 9 items and asks respondents how often, over the

last two weeks, they have been bothered by certain problems, e.g. feeling tired and having

little energy or feeling down, depressed or hopeless (Appendix II). Response options are “not

at all”, “several days”, “more than half the days” and “nearly every day” and the answers are

scored as 0, 1, 2, and 3. The scale provides a 0 to 27 severity score and cut points of 5, 10, 15

and 20 might be interpreted as mild, moderate, moderately severe and severe depression.


Research has shown that PHQ-9 is a reliable and valid measure of depression severity

(Kroenke, Spitzer, & Williams, 2001). The internal reliability was good (Cronbach α = 0.86 -

0.89) and the test-retest reliability was excellent. The results furthermore revealed that there

was a strong association between PHQ-9 scores and functional status, symptom related

difficulty and self-reported sick days and clinic visits, which indicates that construct validity

was good. The PHQ-9 has been translated into Icelandic by Hafrún Kristjánsdóttir, Agnes

Agnarsdóttir, Pétur Tyrfingsson and Jakob Smári (as cited in Valdís Eyja Pálsdóttir, 2007).

The psychometric properties of the Icelandic version have been assessed and the results

revealed that the internal reliability is good (Cronbach α = 0.61 – 0.95) (Valdís Eyja

Pálsdóttir, 2007). The convergent validity of the PHQ-9 is good and that was shown by its

correlations to Beck Depression Inventory-II (r = 0.80) and Clinical Outcomes in Routine

Evaluation – Outcome Measure (CORE-OM) (r = 0.81). The correlations of PHQ-9 to

Quality of Life Scale (QOLS) furthermore proved the divergent validity of the scale (r = -

0.65).

BAI measures anxiety in adults and adolescents (Beck, Epstein, Brown, & Steer, 1988).

The scale consists of 21 questions and respondents are asked to report the extent to which

they have been bothered by each of the symptoms, e.g. numbness or tingling, unsteady and

fear of losing control, over the last week. Each symptom item has four response options: “not

at all”, “mildly - it did not bother me much”, “moderately – it was very unpleasant but I could

stand it” and “Severely – I could barely stand it” and the answers are scored as 0, 1, 2, and 3.

The scale ranges from 0 to 63 and a total score of 0 – 7 is interpreted as a minimal level of

anxiety, 8 – 15 as mild, 16 – 25 as moderate and a total score of 26 – 63 as severe level of

anxiety. The research conducted by Beck et al. showed that BAI has high internal reliability

(Cronbach α = 0.92) and test-retest reliability is also good (r = 0.75) (1988). The convergent

validity of the BAI is also good and that is proved by its moderate correlation with the


revised Hamilton Anxiety Rating Scale (r = 0.51) and mild correlation with the revised

Hamilton Depression Rating Scale (r = 0.25). The results furthermore revealed that the BAI

discriminates anxious diagnostic groups from nonanxious diagnostic groups. The BAI has

been translated into Icelandic and its psychometric properties have been assessed and are

considered satisfactory (Sæmundsson et al., 2011). The results showed that the translated

version of BAI has high internal reliability (Cronbach α = 0.92 - 0.96) and that test-retest

reliability is also good (r = 0.81). The convergent and divergent validity were furthermore

assessed and the results revealed that BAI had higher correlations with the DASS Anxiety

scale than with both the DASS Depression scale and the DASS Stress scale, which supports

good convergent validity and divergent validity (Sæmundsson et al., 2011).

CORE-OM measures participants’ psychological distress (Evans et al., 2002). The scale

consists of 34 items that measure well-being, function, symptoms and risk (Appendix III).

Each item has five response options: “not at all”, “only occasionally”, “sometimes”, “often”

and “all or most of the time” and the answers are scored as 0, 1, 2, 3 and 4. Total scores range

between 0 and 4 and higher scores reflect more severe problems. The research conducted by

Evans et al. revealed that the internal reliability and test-retest reliability of the scale were

good (0.75 – 0.95). The convergent validity was proved by its correlations with seven other

instruments and divergent validity was proved by the large difference between clinical and

non-clinical samples. It has furthermore a good sensitivity to change. The scale has been

translated into Icelandic and its psychometric properties have been evaluated (Kristjánsdóttir

et al., 2013). The Icelandic version has excellent internal reliability (Cronbach α = 0.94),

good test-retest reliability (r = 0.80), acceptable convergent validity and seems to be sensitive

to changes.

QOLS measures quality of life (Burckhardt & Anderson, 2003). It is a 16-item instrument

that measures material and physical well-being, relationships with other people, social,


community and civic activities as well as personal development and fulfillment, recreation

and independence (Appendix IV). Each item has seven response options: “Terrible”,

“Unhappy”, “Mostly Dissatisfied”, “Mixed”, “Satisfied”, “Mostly pleased” and “Delighted”

and the answers are scored from 1 to 7, 1 being terrible and 7 being delighted. The scale is

scored by adding up the score on each item to yield a total score for the instrument. Scores

can range from 16 to 112 whereas higher score indicates higher quality of life and average

total score for a healthy population is about 90. The scale has good internal reliability

(Cronbach α = 0.82 – 0.92) and has high test-retest reliability over 3 weeks (r = 0.78 – 0.84)

(Burckhardt & Anderson, 2003). Convergent validity was proved by its high correlations

with Life Satisfaction Index (r = 0.67 – 0.75). There was furthermore evidence that groups of

patients with persistent condition have lower scores on the scale than healthy adults and

adults with more stable chronic illnesses, which indicates divergent validity. The scale has

been translated into Icelandic and the psychometric properties have been assessed (Ólafur V.

Hrafnsson & Matthías Guðmundsson, 2007). The internal reliability of the translated version

is good (Cronbach α = 0.82 - 0.89) and test-retest reliability is also good (r = 0.72). The scale

has strong positive correlation with Satisfaction With Life Scale, which indicates convergent

validity and its divergent validity was proved by its negative correlation with Beck

Depression Inventory, BAI and Perceived Stress Scale.

MINI is a structured diagnostic interview used to explore 17 mental disorders according to

the diagnostic criteria of the DSM-IV and ICD-10 (Lecrubier et al., 1997; Sheehan et al.,

1998). It was developed for epidemiological studies and multicentre clinical trials. Previous

study has shown that the original scale has excellent reliability, good inter-rater reliability

and good test-retest reliability (Lecrubier et al., 1997). The psychometric properties of the

Icelandic version of the MINI have not been thoroughly studied but one preliminary study

does give some support to its validity (as cited in Kristjánsdóttir et al., 2013).


Procedure

The GAD-7, PHQ-9, BAI, CORE-OM and QOLS were administered to the participants in

the non-clinical sample and the GAD-7 again two weeks later. The participants in this sample

were informed that the participation was optional and they could leave the study at any time,

for any reason. The GPs made clinical evaluations for the patients in the clinical sample and

could refer them to the TCBGT. The referred patients were assessed by a clinical

psychologist in a participation interview, with MINI and psychological scales. If the patients

had more than two disorders and by that fulfilled the MINI criteria a clinical psychologist

gave them a main diagnosis. Patients in the clinical sample filled out GAD-7, PHQ-9 and a

diagnostic specific questionnaire, that was based on main diagnosis evaluation, in diagnostic

interview and every treatment session. Therapy sessions were once a week for six weeks and

lasted for two hours each session.

Permission for the study was obtained from the National Bioethics Committee in Iceland

(VSNb2005090003) and the study was approved by the Icelandic Data Protection Authority

(S2602/2005).

Statistical analysis

All analysis were conducted with SPSS, except the ROC that was conducted with

MedCalc. Test-retest was assessed using the intra-class correlation coefficient (ICC) and the

internal consistency was assessed using Chronbach’s alpha. Calculating the Pearson

correlation coefficient between the GAD-7 and the four other lists, PHQ-9, BAI, CORE-OM

and QOLS, assessed convergent and divergent validity of the GAD-7. The comparison of

patients and participants in the non-clinical sample was determined by independent t-test and

the comparison of patients with or without anxiety disorder diagnosis was determined by

independent t-test as well. Signal detection analysis was conducted for the GAD-7 with

regard to the diagnosis of an anxiety disorder with the MINI. A receiver operating


characteristics (ROC) curve that plots specificity versus sensitivity for every possible cutoff

point was obtained and Youden’s Index (J) was used to evaluate the cutoff point.

Results

Means and Standard Deviations

The mean score for the GAD-7 in the patient sample was 8.60 (SD = 4.77). The mean

score for males was 7.11 (SD = 4.43) and for females 8.90 (SD = 4.79). There was a

statistically significant difference between the genders as determined by independent t-test,

t(219) = 2.104, p = .036.

The mean score in the non-clinical sample for the first administration of the GAD-7 was

1.05 (SD = 2.32). The mean score for males was .55 (SD = 1.57) and for females 1.39 (SD =

2.67). There was a statistically significant difference between the genders as determined by

independent t-test, t(123) = 2.024, p = .045. The mean score for the second administration of

the GAD-7 in the non-clinical sample was 2.00 (SD = 2.89). The mean score for males was

.71 (SD = 1.41) and for females 3.18 (SD = 3.38). There was a statistically significant

difference between the genders as determined by independent t-test, t(71) = 4.010, p = .000.

Internal reliability and test-retest reliability

Internal reliability of the GAD-7 in the clinical sample was α = .867 and in the non-

clinical sample it was α = .863. The two weeks test-retest reliability was assessed in the non-

clinical sample and it was r = .584.

Validity

Pearson correlation coefficient was calculated between the GAD-7 scale and the four other

scales, BAI, CORE-OM, PHQ-9 and QOLS, to assess the convergent and divergent validity

of the GAD-7 scale. This was done both for the non-clinical sample and the clinical sample

(see Table 1).


Table 1. Correlations of the GAD-7 with the BAI, CORE-OM, QOLS and PHQ-9 in both samples BAI CORE-OM QOLS PHQ-9

GAD-7 in the non-clinical sample

.654* (n = 125)

.641* (n = 125)

-.373* (n = 125)

.653* (n = 125)

GAD-7 in the clinical sample

.652* (n = 197)

.709* (n = 204)

-.409* (n = 193)

.631* (n = 211)

Note. * p < .01.

For the non-clinical sample the GAD-7 had highest correlation with BAI indicating good

convergent validity (p < .01) and lower correlation with QOLS supporting good divergent

validity (p < .01). The correlation with PHQ-9 was similar to the correlation of GAD-7 with

BAI. For the clinical sample the GAD-7 had highest correlation with CORE-OM followed by

BAI, indicating convergent validity (p < .01). The scale had lower correlation with QOLS

indicating divergent validity (p < .01). The correlation with PHQ-9 was similar to the

correlation of GAD-7 with BAI.

Comparison of Patients in Clinical-Sample and Participants in Non-Clinical Sample

and Comparison of Patients With or Without Anxiety Disorder Diagnoses

The difference between patients scores and scores of participants in the non-clinical

sample on the GAD-7 was determined by independent t-test. This was done to assess the

discriminant validity of the list as a screener for GAD. The mean score on GAD-7 for

patients in the clinical sample was 8.60 (SD = 4.77) and for the individuals in the non-clinical

sample it was 1.06 (SD = 2.32), the difference between the two groups was statistically

significant (p < .001).

In order to assess the discriminant validity of the GAD-7 for screening for GAD the scores

of patients with diagnosis of anxiety disorder were compared with patients’ scores that did

not have anxiety disorder diagnosis. The mean score on GAD-7 for patients with anxiety

disorder diagnosis was 10.23 (SD = 4.50) and for patients without diagnosis the mean score

was 5.50 (SD = 3.59). The mean score for males with diagnosis was 8.67 (SD = 4.28) and for

females it was 10.44 (SD = 4.51). For patients without anxiety disorder diagnosis the mean


score for males was 5.63 (SD = 4.14) and for females it was 5.46 (SD = 3.43). The difference

between the patient groups, that is with or without anxiety disorder diagnosis, was

statistically significant (p < .001).

Signal Detection Analysis/Receiver Operating Characteristic Analysis

Signal detection analysis was conducted for the GAD-7 in the clinical sample in order to

map the scores on the GAD-7 to the MINI diagnoses. A ROC curve that plots specificity

versus sensitivity for every possible cutoff point was obtained. In all, 147 of 225 patients

were diagnosed with an anxiety disorder. The analysis showed that the GAD-7 discriminates

well between patients diagnosed with anxiety disorder and other patients (Area Under the

Curve (AUC) = .80, 95% confidence interval = .74 - .85). The Yuden’s Index (J) was used to

determine the optimal cutoff value from the ROC analysis and it was obtained for the value

>7 (Youden, 1950). At this point the sensitivity was 66.67% and specificity was 78.21% (see

Table 2).

Table 2. Operating Characteristics of GAD-7 at Different Cutoffs Criterion Sensitivity % 95% CI Specificity % 95% CI

> 3 97.28 93.2 – 99.3 26.92 17.5 – 38.2 > 4 91.84 86.2 – 95.7 46.15 34.8 – 57.8

> 4,67 91.84 86.2 – 95.7 47.44 36.0 – 59.1 > 5 82.31 75.2 – 88.1 56.41 44.7 – 67.6 > 6 75.51 67.7 – 82.2 64.10 52.4 – 74.7 > 7 66.67 58.4 – 74.2 78.21 67.4 – 86.8 > 8 59.86 51.5 – 67.9 83.33 73.2 – 90.8

> 8,17 59.18 50.8 – 67.2 83.33 73.2 – 90.8 > 9 54.42 46.0 – 62.6 87.18 77.7 – 93.7

> 10 51.02 42.7 – 59.3 88.46 79.2 – 94.6 > 11 42.18 34.1 – 50.6 93.59 85.7 – 97.9 > 12 31.29 23.9 – 39.5 96.15 89.2 – 99.2 > 13 25.17 18.4 – 33.0 97.44 91.0 – 99.7 > 14 17.01 11.3 – 24.1 97.44 91.0 – 99.7 > 15 12.24 7.4 – 18.7 98.72 93.1 – 100 > 16 8.16 4.3 – 13.8 100 95.4 - 100


Discussion

The present study assessed for the first time the psychometric properties of the Icelandic

version of the GAD-7 in clinical and non-clinical samples. The results support the

psychometric properties of the Icelandic version of the GAD-7 and the use of it as a

measuring device for GAD.

The means and standard deviations for the participants in the non-clinical sample on the

GAD-7 in this study were slightly lower than the means in the general population in a study

conducted by Löwe et al., where it was 2.97 (2008). This difference could be due to the age

difference as the mean age was higher in the general population in Löwe’s study than in the

non-clinical sample in this study. The age difference could indicate more responsibilities and

burdens among the general population in Löwe’s study and due to that these individuals

might experience increased anxiety symptoms. The means and standard deviations for

patients diagnosed with anxiety disorder were slightly lower in this study than in previous

studies of patients with GAD (14.4 in Spitzer et al., 2006 and 14.18 in Löwe et al. 2008). For

patients without diagnosis of anxiety disorder the mean score in this study was similar to the

mean score in previous studies (5.57 in Löwe et al. 2008 and 4.9 in Spitzer et al. 2006).

Furthermore the mean score for females on the GAD-7 was higher than for males, both in the

clinical sample and in the non-clinical sample, and the difference was statistically significant.

Based on this finding it might be worth considering changing the cut points of the GAD-7 for

males.

The internal reliability of the Icelandic version of the GAD-7 was good, .867 for patients

in the clinical sample and .863 for participants in the non-clinical sample, and that is

comparable to previous studies where the internal reliability of the GAD-7 has proven to be

good (Delgadillo et al., 2012; Donker et al., 2011; Garcia-Campayo et al., 2010; Löwe et al.,

2008; Spitzer et al., 2006). The two weeks test-retest reliability was lower in this study than


in previous research (Spitzer et al., 2006). The reasons for lower test-retest reliability vary

and among them are passage of time, interventions and emotional trauma (Cohen, Swerdlik,

& Sturman, 2013). In the research conducted by Spitzer et al. the participants filled out the

GAD-7 scale within one week from completing the research questionnaire but in this study

the time between the two trials was two weeks (2006). The passage of time might therefore

help explain why the test-retest reliability was lower in this study than in the study conducted

by Spitzer et al. The fact that the second administration of the scale was done during

midterms might as well have effects and lower the test-retest reliability since students

possibly experience increased anxiety symptoms as a consequence of the midterms.

The convergent and divergent validity of the Icelandic version of the GAD-7 were

supported in this study. The correlation coefficient between the GAD-7 and BAI was

relatively high in both samples and that indicates good convergent validity and is similar to

the findings of Spitzer et al. where correlation of GAD-7 with BAI was 0.72 (2006). Lower

correlation of the GAD-7 in both samples with QOLS support good divergent validity as this

questionnaire measures quality of life (Burckhardt & Anderson, 2003). The correlation with

PHQ-9 was similar to the correlation with BAI in both samples and that is in accordance with

previous studies of anxiety and depression. The comorbidity of depressive and anxiety

disorders in addition to the correlation of GAD-7 scores with scores on depression measures

is a well known fact (as cited in Spitzer et al., 2006).

The discriminant validity of the GAD-7 was assessed by examining the difference

between the patients scores and scores of participants in the non-clinical sample. The

difference between the two groups was statistically significant for the scale and the

questionnaire proved to be good and valid screener for GAD. Research has emphasized that it

is important in clinical strategy to have reliable and valid measures of disorders and it is

probably as important that these measuring devices can discriminate between symptoms and


disorders (Delgadillo et al., 2012). The discriminant validity of the GAD-7 for screening for

GAD was assessed as well. This was done by comparing scores of patients diagnosed with

anxiety disorder on GAD-7 to scores of patients that did not have such diagnosis. The results

support the GAD-7 as a screener for GAD since the mean score of patients with anxiety

disorder diagnosis was higher than the scores of patients without anxiety disorder diagnosis

and the difference was statistically significant. This is consistent with previous findings

where the mean score for individuals diagnosed with GAD (mean = 13.96, SD = 4.19) was

higher than the mean score in the control group (mean = 3.54, SD = 3.32), and the difference

between the groups was significant (Garcia-Campayo et al., 2010).

The signal detection analysis revealed that GAD-7 discriminates well between

patients diagnosed with anxiety disorder and patients without such diagnosis. When

developing the scale Spitzer et al. proposed a cutoff value of 10 or greater as a reasonable cut

point for identifying cases of GAD (2006). In this study the cutoff score for the GAD-7 was 7

or greater which is slightly lower than in previous studies where the cutoff score has been 8-

10 or greater (Delgadillo et al., 2012; Kroenke et al., 2007).

The findings in this study are consistent with findings from previous research that state

that GAD-7 is a valid and reliable measuring device. The results of this study support the use

of the Icelandic version of the GAD-7 as a screener for GAD both in the general population

and among patients. The psychometric properties of the Icelandic version are similar to the

properties of the original GAD-7 scale and are considered satisfactory.


References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental

disorders: DSM-5 (5th ed.). Washington, D.C: American Psychiatric Association.

Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring

clinical anxiety: Psychometric properties. Journal of Consulting and Clinical

Psychology, 56(6), 893–897. doi:10.1037/0022-006X.56.6.893

Burckhardt, C. S., & Anderson, K. L. (2003). The Quality of Life Scale (QOLS): Reliability,

Validity, and Utilization. Health and Quality of Life Outcomes, 1(1), 60.

doi:10.1186/1477-7525-1-60

Butcher, J. N., Mineka, S., & Hooley, J. M. (2013). Abnormal psychology. Boston: Pearson.

Cohen, R. J., Swerdlik, M. E., & Sturman, E. (2013). Psychological testing and assessment:

an introduction to tests and measurement. New York, NY: McGraw-Hill.

Delgadillo, J., Payne, S., Gilbody, S., Godfrey, C., Gore, S., Jessop, D., & Dale, V. (2012).

Brief case finding tools for anxiety disorders: Validation of GAD-7 and GAD-2 in

addictions treatment. Drug and Alcohol Dependence, 125(1-2), 37–42.

doi:10.1016/j.drugalcdep.2012.03.011

Donker, T., van Straten, A., Marks, I., & Cuijpers, P. (2011). Quick and easy self-rating of

Generalized Anxiety Disorder: Validity of the Dutch web-based GAD-7, GAD-2 and

GAD-SI. Psychiatry Research, 188(1), 58–64. doi:10.1016/j.psychres.2011.01.016

Evans, C., Connell, J., Barkham, M., Margison, F., McGrath, G., Mellor-Clark, J., & Audin,

K. (2002). Towards a standardised brief outcome measure: psychometric properties

and utility of the CORE—OM. The British Journal of Psychiatry, 180(1), 51–60.

doi:10.1192/bjp.180.1.51

Garcia-Campayo, J., Zamorano, E., Ruiz, M. A., Pardo, A., Perez-Paramo, M., Lopez-

Gomez, V., … Rejas, J. (2010). Cultural adaptation into Spanish of the generalized


anxiety disorder-7 (GAD-7) scale as a screening tool. Health and Quality of Life

Outcomes, 8. doi:10.1186/1477-7525-8-8

Hansell, J. (2005). Abnormal psychology. Hobeken, NJ: Wiley.

Konkan, R., Senormanci, O., Guclu, O., Aydin, E., & Sungur, M. Z. (2013). Validity and

Reliability Study for the Turkish Adaptation of the Generalized Anxiety Disorder-7

(GAD-7)Scale. Noropsikiyatri Arsivi-Archives of Neuropsychiatry, 50(1), 53–58.

doi:10.4274/npa.y6308

Kristjánsdóttir, H., Sigurðsson, B. H., Salkovskis, P., Olason, D., Sigurdsson, E., Evans, C.,

… Sigurðsson, J. F. (2013). Evaluation of the Psychometric Properties of the

Icelandic Version of the Clinical Outcomes in Routine Evaluation-Outcome Measure,

its Transdiagnostic Utility and Cross-Cultural Validation. Clinical Psychology &

Psychotherapy. doi:10.1002/cpp.1874

Kroenke, K., & Spitzer, R. L. (2002). The PHQ-9: A New Depression Diagnostic and

Severity Measure. Psychiatr Ann, 32(9), 1–7.

Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001). The PHQ-9: Validity of a Brief

Depression Severity Measure. Journal of General Internal Medicine, 16(9), 606–613.

doi:10.1046/j.1525-1497.2001.016009606.x

Kroenke, K., Spitzer, R. L., Williams, J. B. W., Monahan, P. O., & Lowe, B. (2007). Anxiety

disorders in primary care: Prevalence, impairment, comorbidity, and detection. Annals

of Internal Medicine, 146(5), 317–325.

Lecrubier, Y., Sheehan, D., Weiller, E., Amorim, P., Bonora, I., Harnett Sheehan, K., …

Dunbar, G. (1997). The Mini International Neuropsychiatric Interview (MINI). A

short diagnostic structured interview: reliability and validity according to the CIDI.

European Psychiatry, 12(5), 224–231. doi:10.1016/S0924-9338(97)83296-8


Löwe, B., Decker, O., Mueller, S., Braehler, E., Schellberg, D., Herzog, W., & Herzberg, P.

Y. (2008). Validation and standardization of the generalized anxiety disorder screener

(GAD-7) in the general population. Medical Care, 46(3), 266–274.

Ólafur V. Hrafnsson & Matthías Guðmundsson. (2007). Próffræðilegir eiginleikar

Lífsgæðakvarðans (QOLS) (unpublished bachelor´s thesis). University of Iceland,

Psychology.

Roy-Byrne, P. P., & Wagner, A. (2004). Primary care perspectives on generalized anxiety

disorder. Journal of Clinical Psychiatry, 65, 20–26.

Sæmundsson, B. R., Þórsdóttir, F., Kristjánsdóttir, H., Ólason, D. Þ., Smári, J., & Sigurðsson,

J. F. (2011). Psychometric properties of the Icelandic version of the Beck Anxiety

Inventory in a clinical and a student population. European Journal of Psychological

Assessment, 27(2), 133–141. doi:10.1027/1015-5759/a000059

Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P., Janavs, J., Weiller, E., …

Dunbar, G. C. (1998). The Mini-International Neuropsychiatric Interview (M.I.N.I.):

the development and validation of a structured diagnostic psychiatric interview for

DSM-IV and ICD-10. The Journal of Clinical Psychiatry, 59 Suppl 20, 22–33;quiz

34–57.

Spitzer, R. L., Kroenke, K., & Williams, J. B. W. (1999). Validation and utility of a self-

report version of PRIME-MD - The PHQ primary care study. Jama-Journal of the

American Medical Association, 282(18), 1737–1744. doi:10.1001/jama.282.18.1737

Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). A brief measure for

assessing generalized anxiety disorder - The GAD-7. Archives of Internal Medicine,

166(10), 1092–1097. doi:10.1001/archinte.166.10.1092

Spitzer, R. L., Williams, J. B. W., Kroenke, K., Linzer, M., Degruy, F., Hahn, S., … Johnson,

J. (1994). Utility of New Procedure for Diagnosing Mental-Disorders in Primary-Care


- the Prime-Md-1000 Study. Jama-Journal of the American Medical Association,

272(22), 1749–1756. doi:10.1001/jama.272.22.1749

Spitzer RL, Williams JW, Gibbon M, & First MB. (1992). The structured clinical interview

for dsm-iii-r (scid): I: history, rationale, and description. Archives of General

Psychiatry, 49(8), 624–629. doi:10.1001/archpsyc.1992.01820080032005

Valdís Eyja Pálsdóttir. (2007). Réttmæti sjálfsmatskvarðans Patient Health Questionnaire

(PHQ) gagnvart geðgreiningarviðtalinu Mini International Neuropsychiatric Interview

(MINI) við að greina geðraskanir hjá heilsugæslusjúklingum (unpublished bachelor´s

thesis). University of Iceland, Psychology.

Youden, W. J. (1950). Index for rating diagnostic tests. Cancer, 3(1), 32–35.

doi:10.1002/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3


Appendix I

GAD - 7

Hversu oft á síðastliðnum 2 vikum hefur þér liðið illa vegna eftirfarandi? Aldrei Nokkra

daga

Oftar en helming daganna

Næstum daglega

1. Verið spennt/-ur á taugum, kvíðin/-n eða hengd/-ur upp á þráð

2. Ekki tekist að bægja frá þér áhyggjum eða hafa stjórn á þeim

3. Haft of miklar áhyggjur af ýmsum hlutum

4. Átt erfitt með að slaka á

5. Verið svo eirðarlaus að þú áttir erfitt með að sitja kyrr

6. Orðið gröm/gramur eða pirruð/pirraður af minnsta tilefni

7. Verið hrædd/-ur eins og eitthvað hræðilegt gæti gerst


Appendix II

PHQ-9

______________________ _____ ________

Hversu oft hefur eftirfarandi vandamál truflað þig síðastliðnar

tvær vikur?

Alls ekki

Nokkra daga

Meira en helming tímans

Nánast

alla daga

a. Lítill áhugi eða gleði við að gera hluti.......…………….....

b. Verið niðurdregin/n dapur/döpur eða vonlaus.……….....

c. Átt erfitt með að sofna eða sofa alla nóttina.......….......

d. Þreyta og orkuleysi....................................……….....…..

e. Lystarleysi eða ofát.......................…................……….….

f. Liðið illa með sjálfan þig eða fundist að þér hafi mistekist eða ekki staðið þig í stykkinu gagnvart sjálfum þér eða fjölskyldu þinni..............................................................

g. Erfiðleikar með einbeitingu við t.d. að lesa blöðin eða horfa á sjónvarp......................................................….

h. Hreyft þig eða talað svo hægt að aðrir hafa tekið eftir þvi? Eða hið gagnstæða – verið svo eirðarlaus eða óróleg(ur) að þú hreyfðir þig mikið meira en venjulega..........……..

i. Hugsað um að það væri betra að þú værir dáin(n) eða hugsað um að skaða þig á einhvern hátt…………..........

Íslensk þýðing með leyfi höfunda: Agnes Agnarsdóttir, Hafrún Kristjánsdóttir, Jakob Smári, Jón Friðrik Sigurðson og Pétur Tyrfingsson


Appendix III

CLINICAL OUTCOMES in ROUTINE EVALUATION ÁRANGURSMAT

MIKILVÆGT – LESIÐ ÞETTA FYRST

Á þessu eyðublaði eru 34 fullyrðingar um hvernig þér hefur gengið SÍÐUSTU VIKUNA. Vinsamlegast lestu hverja staðhæfingu fyrir sig og hugleiddu hversu oft þér leið þannig síðustu vikuna. Merktu svo í þann reit sem á best við svar þitt. Hvernig hefur þér liðið síðustu vikuna?

1 Ég hef verið hræðilega einmana og einangruð/einangraður. ! 0 ! 1 ! 2 ! 3 ! 4 !! F

2 Ég hef verið spennt(ur), kvíðin(n) eða taugaóstyrk(ur). ! 0 ! 1 ! 2 ! 3 ! 4 !! P P

3 Mér fannst ég geta leitað til einhvers eftir stuðningi þegar ég þurfti.

! 4 ! 3 ! 2 ! 1 ! 0 !! F F

4 Ég hef verið sátt(ur) við sjálfa(n) mig. ! 4 ! 3 ! 2 ! 1 ! 0 !! WW

5 Ég hef verið algjörlega orku– og áhugalaus. ! 0 ! 1 ! 2 ! 3 ! 4 !! P P

6 Ég hef beitt aðra líkamlegu ofbeldi. ! 0 ! 1 ! 2 ! 3 ! 4 !! R R

7 Þegar eitthvað hefur gengið illa hef ég getað tekist á við það.

! 4 ! 3 ! 2 ! 1 ! 0 !! F F

8 Óþægindi, verkir eða önnur líkamleg vandamál hafa truflað mig.

! 0 ! 1 ! 2 ! 3 ! 4 !! P P

9 Ég hef hugsað um að skaða sjálfa(n) mig. ! 0 ! 1 ! 2 ! 3 ! 4 !! R R

10 Mér hefur fundist of erfitt að tala við fólk. ! 0 ! 1 ! 2 ! 3 ! 4 !! F F

11 Spenna og kvíði hafa komið í veg fyrir að ég gerði mikilvæga hluti.

! 0 ! 1 ! 2 ! 3 ! 4 !! P P

12 Ég hef verið ánægð(ur) með það sem ég hef gert. ! 4 ! 3 ! 2 ! 1 ! 0 !! F F

13 Óvelkomnar hugsanir og tilfinningar hafa truflað mig. ! 0 ! 1 ! 2 ! 3 ! 4 !! P P

14 Ég hef verið gráti nær. ! 0 ! 1 ! 2 ! 3 ! 4 !! WW

Alls ekki

Stöku sinnum Stundum Oft

Næstum alltaf eða alltaf

Aðeins fyrir

starfsfólk

_____

____________________________________________________ ________________ _________ ___ ___


15 Ég hef fundið fyrir ofsakvíða eða skelfingu. ! 0 ! 1 ! 2 ! 3 ! 4 !! P P

16 Ég hef gert áætlanir um að stytta mér aldur. ! 0 ! 1 ! 2 ! 3 ! 4 !! R R

17 Mér hafa fundist vandamál mín vera yfirþyrmandi. ! 0 ! 1 ! 2 ! 3 ! 4 !! WW

18 Ég hef átt í erfiðleikum með að sofna eða ná að sofa alla nóttina.

! 0 ! 1 ! 2 ! 3 ! 4 !! P P

19 Ég hef fundið fyrir hlýju og væntumþykju til einhvers. ! 4 ! 3 ! 2 ! 1 ! 0 !! F F

20 Ég hef ekki getað hætt að hugsa um vandamál mín. ! 0 ! 1 ! 2 ! 3 ! 4 !! P P

21 Ég hef getað gert flest af því sem ég þarf að gera. ! 4 ! 3 ! 2 ! 1 ! 0 !! F F

22 Ég hef ógnað eða hótað einhverjum. ! 0 ! 1 ! 2 ! 3 ! 4 !! R R

23 Ég hef fundið fyrir vonleysi eða örvæntingu. ! 0 ! 1 ! 2 ! 3 ! 4 !! P P

24 Ég hef hugsað að það væri best að ég væri dáin(n). ! 0 ! 1 ! 2 ! 3 ! 4 !! R R

25 Mér hefur fundist aðrir vera að gagnrýna mig. ! 0 ! 1 ! 2 ! 3 ! 4 !! F F

26 Mér hefur fundist ég ekki eiga vini. ! 0 ! 1 ! 2 ! 3 ! 4 !! F F

27 Ég hef verið óhamingjusöm/samur. ! 0 ! 1 ! 2 ! 3 ! 4 !! P P

28 Óvelkomnar minningar eða hugsanir hafa valdið mér vanlíðan.

! 0 ! 1 ! 2 ! 3 ! 4 !! P P

29 Ég hef verið pirruð/pirraður þegar ég er með öðru fólki. ! 0 ! 1 ! 2 ! 3 ! 4 !! F F

30 Ég hef hugsað um að vandamál mín og erfiðleikar séu sjálfri/sjálfum mér að kenna.

! 0 ! 1 ! 2 ! 3 ! 4 !! P P

31 Ég hef verið bjartsýn(n) á framtíðina. ! 4 ! 3 ! 2 ! 1 ! 0 !! WW

32 Mér hefur tekist það sem ég ætlaði mér. ! 4 ! 3 ! 2 ! 1 ! 0 !! F F

33 Mér hefur fundist annað fólk hafa niðurlægt mig eða fyllt mig skömm.

! 0 ! 1 ! 2 ! 3 ! 4 !! F F

34 Ég hef skaðað mig líkamlega eða stofnað heilsu minni í alvarlega hættu.

! 0 ! 1 ! 2 ! 3 ! 4 !! R R

Alls ekki Stöku sinnum Stundum Oft

Næstum alltaf eða

alltaf

Aðeins fyrir

starfsfólk


Appendix IV

Mat á lífsgæðum

(Quality of Life Scale)

Vinsamlegast skoðaðu hvert tölusett atriði á listanum hér fyrir neðan sem á að lýsa ákveðnum sviðum lífs þíns. Dragðu hring um þá tölu sem lýsir því best hversu ánægð(ur) eða óánægð(ur) þú ert með líf þitt á hverju sviði fyrir sig. Merktu við atriði jafnvel þótt þú sért ekki virk(ur) á því sviði. Við getum verið ánægð/óánægð með að vera ekki virk á einhverju sviði eða ekki í þeim tengslum við fólk sem spurt er um.

Mjög ánægð(ur)

Ánægð(ur)

Frekar ánægð(ur)

Hvorki ánægð(ur)

né óánægð(ur)

Frekar óánægð(ur)

Óánægð(ur)

Mjög óánægð(ur)

1. Efnisleg gæði – heimili, matur, þægindi og fjárhagslegt öryggi

7 6 5 4 3 2 1

2. Heilsa – líkamlegt heilbrigði og hreysti

7 6 5 4 3 2 1

3. Samband við foreldra, systkini og aðra ættingja – tjáning og tengsl, heimsóknir, aðstoð

7 6 5 4 3 2 1

4. Eiga og ala upp börn 7 6 5 4 3 2 1

5. Náin tengsl við maka eða aðra ástvini 7 6 5 4 3 2 1

6. Nánir vinir 7 6 5 4 3 2 1

7. Að hjálpa öðrum og hvetja, bjóða aðstoð og gefa ráð

7 6 5 4 3 2 1

8. Þátttaka í samtökum og opinberum málum (félagsmálum, þjóðmálum)

7 6 5 4 3 2 1

9. Nám – skólanám, aukin skilningur og bætt þekking

7 6 5 4 3 2 1

10.

Að skilja sjálfan mig – þekkja kosti mína og galla – vita um hvað lífið snýst

7 6 5 4 3 2 1

11.

Störf – vinna innan eða utan heimilis 7 6 5 4 3 2 1

12.

Skapandi tjáning 7 6 5 4 3 2 1

13.

Félagslíf – hitta annað fólk, vera virkur, samkvæmi o.s.frv.

7 6 5 4 3 2 1


14.

Lestur, hlusta á tónlist og fara á sýningar

7 6 5 4 3 2 1

15.

Þáttaka í skipulögðu frístundastarfi 7 6 5 4 3 2 1

16.

Sjálfstæði, – að geta séð um mig, bjargað mér og gert hlutina sjálf(ur)

7 6 5 4 3 2 1

Documents

Psychometric Properties of the Icelandic Version of the ...³sa... · spurningalistum, BAI, CORE-OM og QOLS, auk þess að svara GAD-7 í annað sinn. Innri áreiðanleiki var talinn