Anxiety was defined as “a complex cognitive, affective, physiological, and behavioural response system (i.e., threat mode) that is activated when anticipated events or circumstances are deemed to be highly aversive because they are perceived to be unpredictable, uncontrollable events that could potentially threaten the vital interests of the individual” ([1]; p. 5). Anxiety has been proposed as a risk factor for, a consequence of, and a core component of schizophrenia [2–6]. Anxiety is one of the important and common characteristics observed in schizophrenia. Estimated prevalence rates of comorbid anxiety disorders (obsessive-compulsive disorder, social anxiety, and panic disorder) were reported to be significantly elevated in schizophrenia (45.16%) compared to the general population [5]. Meta-analytic findings demonstrated that 38.3% of patients with schizophrenia met diagnostic criteria for at least one anxiety disorder [3]. Studies have also shown increased rates of subclinical anxiety in schizophrenia, such as a lifetime prevalence of panic attacks of 45% [7].
Previous reports suggest that anxiety symptoms seen in schizophrenia have both similar and distinct features with anxiety disorders [8–11]. For instance, individuals with schizophrenia anxiety express certain symptoms and clinical features that are different from those characterizing any of the specified anxiety syndromes [12–16], such as palpitations, trembling, restlessness, hyper-vigilance, and tension. However, most of the literature available on anxiety in schizophrenia is restricted to only focusing on established diagnostic categories, such as obsessive-compulsive disorder, specific phobias, social anxiety disorder, and panic disorder [3, 8, 17]. It is still unknown and unclear whether anxiety that is not differentiated into these categories (i.e., undifferentiated anxiety [2], defined by the DSM-5 as an unspecified anxiety disorder and anxiety disorder not otherwise specified [18]) occurs as commonly as anxiety disorders in schizophrenia, and whether it can be empirically distinguishable from them. The actual rates of undifferentiated anxiety in schizophrenia are unclear. Some existing studies indicate that it is a common clinical problem, with preliminary estimates being of 36% of patients with schizophrenia who display undifferentiated anxiety in the absence of diagnosable anxiety disorder [2].
Overlooking anxiety in schizophrenia may increase the burden of the disease and lead to multiple detrimental consequences for patients, including more severe positive symptoms [17], higher risk of suicide and suicide attempts [16, 19], poor antipsychotic medication management [20], higher risk of relapses [21], increased severity of comorbid medical conditions [15], poorer social functioning [14, 22, 23], and lower quality of life [13]. Other adverse outcomes of anxiety in schizophrenia have also been reported, such as longer and more frequent length of hospitalizations, negative attribution style, substance abuse [16], high rates of co-occurring mental disorders, and increased mental health services use [7]. All these observations highlight that accurately assessing and effectively addressing anxiety in schizophrenia may offer significant clinically relevant benefits for patients. Such assessment needs robust measurement instruments.
Measuring anxiety in schizophrenia
Anxiety in schizophrenia has long been evaluated using inadequate and non-specific measures. Among the first and most widely measures of perceived anxiety symptoms used in schizophrenia patients is the Hamilton Anxiety Rating Scale (HAM-A) [24]. However, this scale presents numerous known drawbacks, including not appropriately measuring worry, which is a core component of anxiety, and measuring symptoms that rather capture depression. A systematic review of psychometric properties of measures assessing anxiety in non-affective psychosis concluded that the existing scales that were examined demonstrated poor performance against standardized quality assessment criteria, and no one showed solid psychometric characteristics or adequate methodological quality [25]. Some measures identified by this review as having acceptable properties for general screening include the Scale of Anxiety Evaluation in Schizophrenia (SAES) [26], the Depression Anxiety Stress Scales (DASS) [27], and the Beck Anxiety Index [28], whereas those suggested as likely to be adequate for assessing specific anxiety disorders/symptoms involve the Yale-Brown Obsessive Compulsive Scale) [29], the Obsessive-Compulsive Inventory [30], the Liebowitz Social Anxiety Scale [31], the DSM-based Generalized Anxiety Disorder Symptoms Severity [6], the Perseverative Thinking Questionnaire [32], and the Psychological Stress Index [33]. An example of measures that have theoretically been designed to evaluate undifferentiated anxiety in the particular schizophrenia population, that is the SAES, was shown to have a broad conceptual scope excluding key expressions of anxiety (e.g., compulsions) and not specifically accounting for anxiety that is expressed within the content of psychotic symptoms (as all its items were taken from existing measures of anxiety) [34]. The latter represents a significant flaw, given that hallucinations and/or delusions with threatening content could be experienced by one patient as intensely terrifying and frightening, and not by another one.
To overcome the abovementioned gaps, Van Staden et al. [35] created and validated a new rating scale, i.e. the Staden Schizophrenia Anxiety Scale (S-SARS), to specifically capture anxiety in patients with schizophrenia. The S-SARS is a clinician-administered scale that was conceptualized and developed to account for both formally diagnosed and undifferentiated anxiety. It is composed of ten items. Five items measure general anxiety (worry and fear, control-related anxiety, psychomotor and cognitive agitation, somatic anxiety, and impairment from anxiety), and five other items measure specific anxiety (obsessive-compulsive anxiety, anxiety attacks, situational anxiety, perceptual anxiety, and persecutory and nihilistic anxiety). The S-SARS was initially used in three published studies [2, 36, 37], and later its psychometric properties were examined by analysing and pooling the data from these studies [35]. Results demonstrated the good reliability and validity of the S-SARS for assessing the undifferentiated and specified anxiety in acute and residual phases of schizophrenia [35]. Later, the good reliability and validity of the S-SARS have been confirmed in a Chinese-speaking sample of patients with schizophrenia [34].
Rationale
Literature on anxiety in patients with schizophrenia of Arab origin is surprisingly scarce, particularly given that expressions of both psychotic disorders [38] and anxiety disorders [39] can be largely shaped by cultural factors. We could find only two previous studies conducted in Lebanon, one used the DASS [40] while the other used the Lebanese Anxiety Scale-10 [41]. It is of note that in the two studies anxiety was not the main focus. The present study proposes to complement the literature by examining the psychometric properties of an Arabic translation of the S-SARS in a sample of chronic, remitted patients with schizophrenia from Lebanon. It has been hypothesized that the Arabic version of the scale will show good factorial validity, high internal consistency, as well as appropriate convergent and concurrent validity against measures of anxiety (i.e., the Generalized Anxiety Disorder 7Item Scale, GAD-7), depression and general functioning. Therefore, the S-SARS is anticipated to be a useful tool for clinicians and researchers working in Arabic-speaking settings for accurately identifying anxiety in schizophrenia. As the Arabic version of the GAD-7 has not been previously validated in an Arabic-speaking clinical population of patients with schizophrenia, this study had as a secondary objective to examine the psychometric properties of this scale before its use.