Regensburg Insomnia Scale (RIS): a new short rating scale for the assessment of psychological symptoms and sleep in insomnia; Study design: development and validation of a new short self-rating scale in a sample of 218 patients suffering from insomnia and 94 healthy controls

Insomnia is a frequent disorder with prevalences varying between 2% [1] and 48% [2] depending on the diagnostic criteria used. It is often associated with psychiatric and organic diseases and produces costs due to accidents and absentieesm [3‐5]. Several tools for measuring insomnia have been developed [6‐9] reflecting the growing interest and need for assessment instruments in insomnia research. However, there still is a need for a short scale measuring the psychological aspects of insomnia together with quantitative aspects of sleep.

Why measure the psychological aspects of insomnia? The reason is that there is growing evidence that psychological symptoms are prominent, perpetuating factors for disturbed sleep [10, 11], which can be best seen in psychophysiological insomnia [12] (PI). Psychological aspects such as sleep related worries and rumination, dysfunctional attitudes toward sleep [13], distorted sleep perception [14] and hyperarousal [15] have been found and investigated thoroughly. It is assumed that insomnia-specific hyperarousal [16] is fuelled by dysfunctional beliefs (such as “I have to sleep at least 8 hours per night”) and by the anticipating the negative consequences of sleep loss. Thus, while trying to fall asleep, enhanced self observation occurs, which in return has an impairing effect on sleep onset [17]. The relevance of psychological aspects in the continuance of chronic insomnia is increasingly being recognized and specific psychotherapeutic treatments such as Cognitive Behaviour Therapy for Insomnia (CBT-I) have been developed and tested [18‐20]. The psychological impact makes it plausible, that PI is comorbid in other sleep disorders such as sleep apnea [21] or Periodic Leg Movement in sleep [22] and the co-existence explains why sleep will not improve by treating the organic disorder [21] alone. In order to assess the psychophysiological components in sleep disorders, a short scale would be useful in clinical routine.

Indeed, there are already a variety of scales assessing psychological aspects of insomnia. For measuring the state of cognitive and somatic arousal there is the Pre-Sleep Arousal Scale [23], the Presleep Cognitive Activity Questionnaire [24] and the Arousal Level as Present State [25]. The Metacognitions Questionnaire-Insomnia (MSQ-I) [26] and the Dysfunctional and Attitudes about Sleep scale [27] have been developed to measure insomnia-specific worrisome thoughts and beliefs. The Thought Control Questionnaire-Insomnia revised scale [28] measures the attempts to control unwanted thoughts in insomnia. The Utility of Presleep Worry Questionnaire (UPWQ) was developed to investigate the utility of presleep worry in insomnia [29] as well as the Sleep Disturbance Questionnaire [30]. The Monitoring for Sleep-Related Threat [31] measures the stimuli that might hinder falling asleep. The Glasgow Sleep Effort Scale was designed to measure the attempt to control the process of falling asleep [32]. All these scales cover psychological aspects of insomnia, however they are too specific for clinical routine, since they each just measure certain symptoms, such as arousal state or presleep worrying.

The Insomnia Severity Index (ISI) [6] is the most established instrument for measuring insomnia symptoms, however there is only one out of 7 items that targets the psychological symptoms, namely being worried about the current sleep pattern. Violani’s Insomnia Scale [9] covers different items such as medication intake, snoring, consulting a doctor, in addition to sleep related items, however this scale does not address clinical psychophysiological symptoms of insomnia and was rather designed for the pre-screening of subjects with insomnia complaints.

A short scale covering quantitative aspects of sleep as well as insomnia specific symptoms is still lacking. Thus we aimed at developing a short new instrument for the assessment of the typical psychological aspects of PI that can be especially used for the evaluation of therapeutic interventions. This scale was validated using a large sample of insomnia patients and healthy controls. Moreover we evaluated the scale’s sensitivity to therapeutic changes in patients with PI who were treated with CBT-I in our hospital. The Regensburg Insomnia Scale (RIS) was administered before and 6 months after completion of therapy. The study was approved by the Ethics Committee of the University of Regensburg.

The RIS is a new self-rating scale with ten items, developed to measure symptoms of PI with special emphasis on psychological symptoms.

The scale covers the characteristic cognitive, emotional and behavioural aspects of PI. These items are essential target symptoms in CBT-I [36, 37]. With an average time for completion and evaluation of less than three minutes, it is a highly efficient tool both for research and clinical practice. The RIS discriminates well between PI and healthy controls. We propose to consider a score from 0–12 points as normal and a higher score as indicative of PI symptoms.

The relatively high sensitivity and specificity values of our scale may be explained by the fact that patients were recruited from a tertiary referral center with many insomnia patients showing high symptom load and a severe degree of insomnia. This is illustrated by the finding that the sensitivity and specificity of the PSQI in our samples were similarly high and even higher than in the original study [35]. Final conclusions about the sensitivity and specificity of the RIS would therefore require additional studies in independent samples. However, it should be noted that the RIS has neither been designed nor validated as a tool for the diagnosis of PI or for differentiating between subtypes of insomnia.

Component analysis revealed that the RIS differentially assesses four factors. Besides “sleep depth” and quantitative aspects of sleep, the factor “fearfully focussing on insomnia” is especially interesting for an insomnia scale. We consider that the combination of psychological aspects that characterize psychophysiological insomnia as well as qualitative and quantitative aspects of sleep all represented in one short scale is the innovative aspect of this scale when compared to existing insomnia scales. Although this multifaceted structure of the RIS indicates its potential for differential assessment of different dimensions of insomnia (sleep quality, sleep quantity, focus on insomnia and medication intake) it should be noted that the RIS has been developed and validated as a single scale and should also be primarily used so.

The RIS was sensitive for detecting improvements after CBT-I in sleep parameters and target symptoms. Both the PSQI and RIS were comparable in measuring effects of CBT-I on sleep latency, sleep duration and hypnotic intake. In addition, the RIS detected effects in target symptoms of CBT-I, especially in thinking about sleep. This is an important result, since modification of dysfunctional beliefs is one of the central aspects of CBT-I [37]. Effects were also seen in sleep-related fear, indicating the sensitivity of the RIS to the emotional aspects of PI. The changes observed in the perception of sleeplessness suggest that the RIS also covers sleep misperception, which is regarded as one of the central aspects of PI [14, 38] . Thus similar to the ISI [6] and the Dysfunctional Beliefs and Attitudes about Sleep Scale [27], the RIS is sensitive to psychotherapy-specific target symptoms and is therefore proposed as a complementary scale with a special focus on psychological symptoms of PI.

Beyond measuring severity of PI, RIS (and especially its different dimensions) may be a useful instrument in disentangling the psychological symptoms from the symptoms of the organic sleep disorders. In a recently published study, we showed that psychophysiological insomnia-specific symptoms contribute to a less compliant attitude towards a treatment of sleep apnea with continuous positive airway pressure [39]. This study was done with the RIS and showed new aspects in comparison to another study that was performed with the ISI earlier [40]. While Nuygen et al., could not see any impact of insomnia symptoms on CPAP compliance with the ISI, two items in the RIS (“I feel that I have not slept all night” and “I am afraid to got to bed because of my disturbed sleep”) specifically correlated with compliance [39].

There are limiting factors in this study design. In this study the RIS has been validated using a sample of insomnia patients with a broad spectrum of severity, ranging from outpatients to patients who received inpatient CBT-I. This is reflected by the broad distribution of the scores (Figure 1). Nevertheless, a potential selection bias cannot be excluded since the sample stems from a specialized tertiary referral centre. Therefore evaluation of the RIS in other samples is strongly encouraged. For the English version normative studies are necessary. Further studies should also investigate the relation to objective sleep data and the comparability to other short insomnia scales recently published. Also the sensitivity of the RIS for other therapeutic interventions such as pharmacotherapy should be addressed. In addition it was not validated against objective measures such as actigrapy or polysomnography. However, since this scale focuses on psychological symptoms and since a wide range in objective sleep parameters in insomnia patients are known [41], we consider this rather a minor problem. Another problem is that this scale was not cross validated against a current insomnia scale. We chose the PSQI because in the evaluation period of the RIS there have been a lot more studies done on insomnia patients with the PSQI rather than the ISI. Furthermore, it was not our purpose to design an alternative to ISI, but rather a new measurement for psychophysiological symptoms and sleep parameters in a short scale with good feasibility.

The RIS is a short, economic and valid instrument for measuring psychological and physiological aspects of PI. It specifically detects changes in target symptoms of CBT-I typical for PI and thus represents an especially well-suited instrument for assessing treatment effects of insomnia specific CBT in both research and clinical use.