The SPAI-18, a brief version of the Social Phobia and Anxiety Inventory: Reliability and validity in clinically referred and non-referred samples

https://doi.org/10.1016/j.janxdis.2013.05.003Get rights and content

Highlights

  • Social anxiety can now be measured in about 8 min with the SPAI-18.

  • The SPAI-18 demonstrates excellent internal consistency and high convergent validity.

  • The SPAI-18 has good to excellent diagnostic properties.

  • The SPAI-18 provides information on various aspects of social anxiety including avoidance.

  • The SPAI-18 is highly suitable for screening purposes in clinical practise.

Abstract

We developed a new version of the Social Phobia and Anxiety Inventory (SPAI) in order to have a brief instrument for measuring social anxiety and social anxiety disorder (SAD) with a strong conceptual foundation. In the construction phase, a set of items representing 5 core aspects of social anxiety was selected by a panel of social anxiety experts. The selected item pool was validated using factor analysis, reliability analysis, and diagnostic analysis in a sample of healthy participants (N = 188) and a sample of clinically referred participants diagnosed with SAD (N = 98). This procedure resulted in an abbreviated version of the Social Phobia Subscale of the SPAI consisting of 18 items (i.e. the SPAI-18), which correlated strongly with the Social Phobia Subscale of the original SPAI (both groups r = .98). Internal consistency and diagnostic characteristics using a clinical cut-off score > 48 were good to excellent (Cronbach's alpha healthy group = .93; patient group = .91; sensitivity: .94; specificity: .88). The SPAI-18 was further validated in a community sample of parents-to-be without SAD (N = 237) and with SAD (N = 65). Internal consistency was again excellent (both groups Cronbach's alpha = .93) and a screening cut-off of >36 proved to result in good sensitivity and specificity. The SPAI-18 also correlated strongly with other social anxiety instruments, supporting convergent validity. In sum, the SPAI-18 is a psychometrically sound instrument with good screening capacity for social anxiety disorder in clinical as well as community samples.

Introduction

Social anxiety is characterised by fear of situations in which a person feels exposed to and possibly judged by others, and fears behaving in an improper or embarrassing way. Social anxiety disorder (SAD), also known as social phobia, is one of the most prevalent and most unsettling mental disorders (Furmark, 2002, Stein and Stein, 2008). For clinical and research purposes, various assessments tools have been developed to measure social anxiety and/or SAD, including short screening instruments (e.g. mini Social Phobia Inventory (mini SPIN) (Connor, Kobak, Churchill, Katzelnick, & Davidson, 2001)), diagnostic interviews (e.g. Anxiety Disorder Interview Schedule (ADIS) (Di Nardo, Brown, & Barlow, 1994)), and self-report measures. A well-known self report measure that has a history of extensive psychometric evaluation is the Social Phobia and Anxiety Inventory (SPAI), which was developed by Turner, Stanley, Beidel, and Bond (1989). The SPAI assesses an extensive range of social anxiety symptoms and has proven to possess excellent psychometric properties (e.g. Beidel et al., 1989, Turner et al., 1996, Turner et al., 1989). For our longitudinal study “Social development of children”, which examines the contribution of predisposition and environmental factors to the development of social anxiety during early childhood, we needed a short instrument to screen parents-to-be on SAD, while preserving assessment of the subjective, behavioural, cognitive, and somatic aspects of social anxiety. We preferred to use the SPAI, because of its conceptual foundation and psychometric quality. A drawback, however, was that completion of the SPAI takes about 20 min. Hence, we chose to construct an abbreviated version of the SPAI.

The SPAI is used in healthy and clinical samples to assess the extent of social anxiety on a continuum (e.g. Beidel, Turner, & Cooley, 1993), for screening of SAD using a clinical cut-off score (e.g. Voncken, Bögels, & de Vries, 2003), and as an outcome measure in for example treatment effect studies (e.g. Beidel et al., 1993, Bögels, 2006, García-López et al., 2005). The SPAI has been translated into various languages and good psychometric characteristics were again established (e.g. Bögels and Reith, 1999, García-López et al., 2001, Olivares et al., 2004, Olivares et al., 1999). The SPAI has also demonstrated good discriminative properties as well as sensitivity to clinical change (Beidel et al., 1993, García-López et al., 2005, Peters, 2000).

Also in comparison to other social anxiety questionnaires such as the Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1998), the Social Phobia Scale (SPS; Mattick & Clarke, 1998), and the Social Phobia Inventory (SPIN; Connor, Davidson, Churchill, Sherwood, & Weisler, 2000), the SPAI stands out. First, with regard to content validity, the SPAI examines a broader range of clinically relevant social anxiety symptoms, including subjective, behavioural, cognitive, and somatic characteristics, while the SIAS and the SPS remain limited to assessing cognitive and subjective aspects, and the SPIN to subjective, behavioural, and somatic symptoms. Second, also psychometrically, the SPAI demonstrates more favourable characteristics than the SIAS, the SPS, and the SPIN. For example, the SPAI has better discriminative quality (e.g. Bögels and Reith, 1999, Peters, 2000) as compared to the SIAS and the SPS (e.g. Heinrichs et al., 2002) and to the SPIN (Connor et al., 2000). Furthermore, although the internal consistency of the SIAS and SPS (Heinrichs et al., 2002, Mattick and Clarke, 1998) and the SPIN (Connor et al., 2000) are good, the internal consistency of the SPAI is even higher (e.g. Bögels and Reith, 1999, Olivares et al., 1999, Osman et al., 1996).

The SPAI includes two subscales: (1) Social Phobia and (2) Agoraphobia (Turner et al., 1989). The former measures characteristics of social anxiety, and the latter assesses fear in situations that are typically associated with agoraphobia, for instance fear for crowds and waiting in lines. The developers added the Agoraphobia subscale to differentiate between social anxiety and anxiety that accompanies agoraphobia. To obtain the SPAI-total score, the Agoraphobia Subscale Score should be subtracted from the Social Phobia Subscale Score. Good, or even superior, results in terms of concurrent and discriminant validity are, however, also obtained with the Social Phobia Subscale Score alone (e.g. Bögels and Reith, 1999, García-López et al., 2001, Herbert et al., 1991, Olivares et al., 2004, Roberson-Nay et al., 2007). For our purpose to obtain a short instrument with good screening capacity for social anxiety, we reasoned that by leaving out the agoraphobia subscale we would probably still reach our main objectives, that is, a short questionnaire with excellent sensitivity and at least adequate specificity. Consequently, we left out the items on agoraphobia.

Roberson-Nay et al. (2007) and Garcia-Lopez, Hidalgo, Beidel, Olivares, and Turner (2008) have developed short versions of the SPAI consisting of 23 items (in English, for adults) and 16 items (in Spanish, for adolescents), respectively, using thorough and sound statistical techniques including item-response analyses. Despite these thorough methods, their procedures have one weakness, namely, the original five-factor model underlying the Social Phobia Subscale was not taken into account for item selection. Since we wanted to ensure capturing the complete social anxiety construct, we strived to retain the theoretically sound five-factor structure of the Social Phobia Subscale, by selecting items representing the various aspects of social anxiety, that is: (a) distress in social situations (individual and group interactions), (b) anxiety when being the centre of attention, (c) avoidance, (d) cognitive symptoms, and (e) somatic reactions (see Turner et al., 1989).

In sum, the objective of this study was to develop a psychometrically sound short instrument to detect social anxiety disorder that can also be used to discriminate between individuals on the social anxiety continuum, and that covers subjective, behavioural, cognitive, and somatic domains of social anxiety. The development of the abbreviated SPAI consisted of two phases: (I) construction and initial validation; and (II) secondary validation using a new sample. In phase I, a panel of experts on social anxiety (AS, MV, MW, PE) selected the items from the original SPAI that referred to the core of social anxiety (disorder) while adhering to the thoroughly formulated conceptual background. Furthermore, we performed a factor analysis and reliability analyses, and assessed the criterion validity on data assembled in a healthy sample and in a sample of clinic-referred patients with SAD in which the original SPAI was administered. In phase II, we validated the abbreviated SPAI by conducting reliability analyses, assessing convergent and divergent validity, distributional features, and diagnostic qualities in a community sample of parents-to-be with and without SAD.

Section snippets

Participants

For the initial development and validation of the abbreviated SPAI, datasets of three studies on social anxiety that were conducted in the Netherlands using highly similar procedures were merged. Taken together this sample (sample A) consisted of 188 patients with SAD and 98 healthy participants. Data coming from the first study (Bögels & Reith, 1999) consisted of 92 clinic-referred patients with SAD diagnosed using the Structured Clinical Interview for DSM-IV (SCID) I (First et al., 1996,

Participants

For the secondary validation of the abbreviated SPAI, data were used from our longitudinal study on the development of social anxiety in young children in a sample of 152 parent couples coming from the general population (sample B). Information on recruitment, selection of participants, and ethical procedures is provided in Majdandžić, de Vente, and Bögels (2010) and De Vente, Majdandžić, Colonnesi, and Bögels (2011). Information of 2 participants was missing on the diagnostic status, assessed

Discussion

The SPAI-18 is a fundamentally solid, short version of the SPAI with good psychometric and diagnostic qualities. Despite the considerable reduction of the number of items, the correlation between the SPAI-18 and the Social Phobia Subscale of the original SPAI is very high. The SPAI-18 also has good distributional features, high internal consistency, even on a factor scale level, good convergent and divergent validity, excellent test–retest reliability, and good sensitivity and specificity.

Acknowledgements

This study was funded by an Innovation Research Vidi NWO grant, number 452-05-345, to the last author. The authors gratefully acknowledge Dr. Agnes Scholing (University of Amsterdam), Prof. Dr. Michiel Westenberg (Leiden University), and Prof. Dr. Paul Emmelkamp (University of Amsterdam) for their expert judgements and selection of the items of the SPAI. We also like to thank Sarah de Schutter, Annefleur Visscher, Jasmijn Rahder, and Hester Verwey for their work in the recruitment and

References (35)

  • S.M. Bögels et al.

    Mindfulness training versus cognitive therapy for social anxiety disorder

    (2013)
  • K.M. Connor et al.

    Psychometric properties of the Social Phobia Inventory (SPIN)

    British Journal of Psychiatry

    (2000)
  • K.M. Connor et al.

    Mini-SPIN: a brief screening assessment for generalized social anxiety disorder

    Depression and Anxiety

    (2001)
  • W. De Vente et al.

    Intergenerational transmission of social anxiety: the role of paternal and maternal fear of negative child evaluation and parenting behaviour

    Journal of Experimental Psychopathology

    (2011)
  • P.A. Di Nardo et al.

    Anxiety Disorders Interview Schedule for DSM-IV. Lifetime version (ADIS-IV-L)

    (1994)
  • M.B. First et al.

    Structured Clinical Interview for DSM-IV Axis I Disorders, Patient edition (SCID-I/P, version 2.0)

    (1996)
  • T. Furmark

    Social phobia: overview of community surveys

    Acta Psychiatrica Scandinavica

    (2002)
  • Cited by (15)

    • Context matters: The role of subjective arousal during Attentional Bias Modification targeting socially anxious students

      2020, Journal of Behavior Therapy and Experimental Psychiatry
      Citation Excerpt :

      The final sample consisted of 79 participants (Mage = 21.92, SD = 2.55; 75.9% female), of which one participant had missing data for the bias assessment of session 2 due to a technical error, one participant dropped out after the first session, and three participants dropped out during the stress reactivity task. The final sample had an average SPAI-18 score of 62.79 (SD = 12.14; range = 38.45–93.15) indicating that the sample had subclinical to clinical levels of social anxiety (de Vente et al., 2014). Social anxiety was assessed with the brief version of the Social Phobia and Anxiety Inventory (SPAI-18; de Vente et al., 2014).

    • Symptoms of social anxiety disorder and major depressive disorder: A network perspective

      2019, Journal of Affective Disorders
      Citation Excerpt :

      Two items were combined because we wanted to capture both interaction and performance fears given that these are considered to be the two broad types of fears for individuals with SAD. We chose these particular items because they are included in a short version of the SPAI (the SPAI-18) that has strong psychometric properties and have been shown to load strongly on correlated factors representing distress in social situations and focus of attention in previous factor analyses (Osman et al., 1995; de Vente et al., 2013). For the social avoidance node, we selected item 7 from the SPAI (I feel so anxious about attending social gatherings that I avoid these situations) because it was included as one of two items that loaded on an avoidance and escape factor in analyses of the SPAI-18 (de Vente et al., 2013) and had the stronger face validity of these two items.

    View all citing articles on Scopus

    The editorial responsibilities for this manuscript were handled by B. Christopher Frueh, Ph.D.

    View full text