Insomnia and Anxiety (Series in Anxiety and Related Disorders) (6 page)

PSG are the overestimation of time spent awake and the underestimation of the time

spent sleeping (Coates & Thoresen, 1979; Means, Edinger, Glenn, & Fins, 2003).

Nonetheless, sleep logs are considered quintessential to insomnia assessment

(Buysse et al., 2006; Sateia, 2002).

18

2 Considerations for Assessment

Clinical Interview

Both structured and semi-structured clinical interviews are commonly used in

clinical and research practice. To arrive at a diagnosis and to develop a case formula-

tion for treatment, the interview focuses on the etiologic (i.e., cognitive and behav-

ioral perpetuating) factors in the insomnia. Thus, the interview will elicit the details

and history of the complaint as well as the history of any possible cooccurring medi-

cal or mental health issue. It is often helpful to attempt to develop a timeline of each

condition to attempt to understand the degree to which the sleep and comorbid con-

ditions are independent, interactive, or dependent. Interviews tend to cover these

major areas: the nature and history of the sleep complaint, current stressors (includ-

ing relationship discord, financial strain, or environmental factors such as a loud or

unsafe sleeping environment), presence of any cardinal symptoms of another sleep

disorder (e.g., loud snoring, a tendency to fall asleep involuntarily, leg twitching,

restless leg symptoms), medical and psychiatric history (including medication use,

surgeries, allergies, exposure to toxins, or any recent change in reproductive status),

current sleep habits (including the presence of shift work or frequent time zone

travel, use of sleep-interfering substances such as caffeine, cigarettes, alcohol), and

treatment history. Information about or from their current bed partner also can be

helpful. For example, the bed partner may exhibit loud snoring, which may be dis-

ruptive to the patient’s sleep. The bed partner may also be helpful in unexpected

ways. For example, a patient was complaining of rather spectacular sleep deprivation

(e.g., she complained that she had not slept in the past 4 years) but she lacked any

appearance of sleepiness and was quite functional during the day. When the husband

was asked about his wife’s sleep problem he said that the problem was her snoring.

The wife had not reported that they were sleeping in separate rooms because of his

complaints about her snoring. Her history of complaints led to an overnight study

that revealed moderate sleep apnea. She also had insomnia and a focus in cognitive

therapy was to examine the anxiety-producing consequence of her belief that she did

not sleep at all (when in actuality she was sleeping, as evidenced by her snoring).

She modified this belief to a more accurate and helpful realization that her sleep was

lightened by the breathing disruptions, and she did indeed sleep.

In addition to an unstructured clinical interview, there are several site-specific

sleep disorder interviews or published semi-structured interviews for insomnia

(Savard & Morin, 2002; Spielman & Anderson, 1999) useful for guiding the prac-

titioner through diagnostic criteria for sleep disorders including insomnia.

Self-report Measurement

Global Sleep Symptom Questionnaires

Arguably, the two most common self-report symptom questionnaires are the

Insomnia Severity Index (ISI) (Morin, 1993) and the Pittsburgh Sleep Quality Index –

PSQI (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). The ISI is a 7-item

Self-report Measurement

19

questionnaire of subjective insomnia symptom severity. Daytime and nighttime

insomnia symptoms are rated using a 5-point (0–4) Likert scale. These symptoms

include: difficulties falling asleep and/or staying asleep, waking too early in the

morning; sleep dissatisfaction; degree of impairment with daytime functioning;

degree to which impairments are noticeable; and distress or concern about insom-

nia. Morin and colleagues suggest the following ranges for interpretation of clinical

significance: 0–7 (no clinical insomnia), 8–14 (sub-threshold insomnia), 15–21

(insomnia of moderate severity), and 22–28 (severe insomnia). There is good reli-

ability and validity (using both sleep logs and electronic sleep recordings) (Bastien,

Vallières, & Morin, 2001). It is a recommended assessment tool for insomnia

research (Buysse et al., 2006), and its quick administration time makes it useful for

clinical use too.

Another recommended measure for standard insomnia assessment (Buysse

et al., 2006) is the Pittsburgh Sleep Quality Index – PSQI (Buysse et al., 1989).

While the ISI is insomnia-specific, the PSQI is a more global measure of sleep

disturbance across sleep disorders. It is a retrospective measure (over the past

month) of sleep onset latency, sleep duration, sleep efficiency (i.e., the proportion

of time in bed that is actually spent asleep), sleep quality, disturbances to sleep,

medication use, and daytime dysfunction. Out of a possible total score that ranges

from 0 to 21, a PSQI score of >5 appears to discriminate those with insomnia from

good sleepers (Buysse et al., 1989). As such, a post-treatment PSQI score <5 has

been used in some studies as indicating insomnia remission. While it is widely used

and has good psychometrics, we have reported that elevated levels of anxiety may

contribute to PSQI score elevations in those with comorbid disorders (Carney,

Edinger, Krystal, Stepanski & Kirby, 2006). Thus, it may be prudent to interpret

PSQI scores with caution in the presence of significant anxiety.

Cognitive Insomnia Questionnaires

The Dysfunctional Beliefs and Attitudes about Sleep Questionnaire – DBAS

(Morin, 1993) is a cognitive measure to assess problematic levels of unhelpful

beliefs about sleep. The most current version is 16 items (Morin) wherein respon-

dents rate the degree to which they believe particular statements about sleep. Both

the original 30-item version and DBAS-16 have acceptable levels of internal con-

sistency (Cronbach’s alpha values >0.80) (Morin, 1993; Morin, Vallières, & Ivers,

2007). The DBAS discriminates between good and poor sleepers and is responsive

to changes in beliefs resulting from cognitive-behavioral therapy for insomnia

(Carney & Edinger, 2006). Responses on specific DBAS items can also be used in

therapy to orient patients to particular unhelpful beliefs and to modify the veracity

of belief in them.

The Sleep Self-Efficacy Scale (SES) (Lacks, 1987) is a 9-item measure of one’s

level of confidence in carrying out particular sleep-related behaviors. Insomnia is

often characterized by thoughts of helplessness (Morin, 1993), so it can be a worth-

while clinical enterprise to determine the level of self-efficacy/agency one has with

20

2 Considerations for Assessment

regards to sleep. The SES has been used in several insomnia trials and has been

shown to improve (i.e., one becomes more confident in the ability to engage in

effective sleep behaviors) with sleep-related improvements (Carney & Edinger,

2006) and to predict response to CBT for insomnia (Edinger et al., 2009). Another

potentially useful measure is the Glasgow Sleep Effort Scale (Broomfield & Espie,

2005). This scale is a measure of sleep-related effort with promising initial psycho-

metric support (Broomfield & Espie, 2005). While further studies are needed, the

concept of sleep effort is a useful one, as it purportedly underlies maladaptive sleep

beliefs (Espie, Broomfield, MacMahon, Macphee, & Taylor, 2006).

Behavioral Insomnia Questionnaires

The Sleep Hygiene Practice Scale (SHAPS) (Lacks, 1987) is a widely used mea-

sure for the presence of sleep-disruptive behaviors such as taking naps, or exercis-

ing strenuously within 2 h of bedtime. While it enjoys frequent usage, the SHAPS

does not appear to have particularly strong internal consistency (Lacks, 1987) and

studies establishing its validity are currently lacking. A lesser known but initially

psychometrically promising tool may be the Sleep Hygiene Index (Mastin, Bryson,

& Corwyn, 2006).

In addition to sleep hygiene behaviors, it may also be important to assess the

presence of safety behaviors. Safety behaviors are those behaviors that are used to

avoid an unwanted experience. In insomnia, an example of a safety behavior would

be consuming alcohol when having difficulty sleeping. One helpful tool in this

regard (i.e., to assess unhelpful safety-related sleep behaviors) is the Sleep-Related

Behaviors Questionnaire (SBRQ) (Ree & Harvey, 2004). This measure was derived

from Harvey’s Cognitive Model (2002) that asserts the safety behaviors that per-

petuate sleep problems – an observation that has been shown experimentally too

(Harvey, 2002).

Daytime Insomnia Symptom Questionnaires

One final issue to consider in the assessment of sleep is the measurement of day-

time impairment. The ISI is useful in that one of the items specifically queries

daytime insomnia symptoms across the range of cognitive, mood, functioning

domains. Additionally, one of the most frequently assessed daytime areas is fatigue.

People with insomnia often complain of fatigue. The Fatigue Severity Scale (FSS)

(Krupp, LaRocca, Muir-Nash, & Sternberg, 1989) is a measure of the severity of

fatigue symptoms. Like the DBAS the total FSS score is a mean-item score of the

responses on the 9 items; a score above 3 is indicative of significant fatigue. While

there are many more comprehensive measures of fatigue available (e.g., the

Assessment of Anxiety

21

Multidimensional Fatigue Inventory), the FSS is brief and has many studies that

establish its strong psychometric properties in those with sleep problems (Krupp

et al., 1989; Krupp, Jandorf, Coyle, & Mendelson, 1993; Lichstein, Means,

Noeb, & Aguillard, 1997).

When dealing with sleep disorders, it is useful to distinguish fatigue from clini-

cally significant sleepiness. This is because sleepiness is often associated with

disorders other than insomnia such as sleep apnea, narcolepsy, or periodic limb

movement disorder. Whereas people with insomnia feel very tired (e.g., fatigued);

they usually do not have clinical levels of sleepiness. Sleepiness is characterized by

the propensity to fall asleep unintentionally, quickly and frequently when given the

opportunity. The widely used Epworth Sleepiness Scale (ESS) (Johns, 1991), is an

8-item self-report questionnaire designed to assess the propensity to fall asleep in

situations such as while driving, watching TV, or sitting and talking to someone.

Respondents rate how likely they would be to fall asleep in these situations using

a 4-point rating scale (0 = “would never doze” to 3 = “high chance of dozing”).

A score of 10 or greater is considered to indicate clinically significant daytime sleepi-

ness. The ESS is a common tool in sleep assessment with good reliability (Johns,

1991) and validity (i.e., strong correlation with objective tests of daytime sleepiness

(Johns, 1991). The most common objective test of sleepiness is the Multiple Sleep

Latency Test (MSLT). The MSLT is conducted at a sleep laboratory and involves

PSG data collection during five 20-min nap opportunities spaced 2 h apart through-

out the day. The sleep onset latency is averaged over the course of the 5 naps to

determine sleepiness. If someone falls asleep within 10 min or less, the person is

regarded as objectively sleepy. In addition to the assessment of sleep and medical

history, it is important to assess for psychiatric factors as well.

Assessment of Anxiety

Those who work with people with sleep disorders assess for a range of psychopa-

thology in addition to the sleep disorder. The diagnosis of insomnia requires that

another disorder cannot better account for the insomnia symptoms, thus it is impor-

tant to understand what other conditions could be causing or affecting the insomnia.

This information is also important in the treatment of insomnia as specific anxiety-

related strategies may need to be added or emphasized in the CBT insomnia treat-

ment package. There are a variety of tools available to assess for general

psychopathology, including semi-structured interviews that assess the range of pos-

sible Axis I disorders (e.g., the Structured Clinical Interview for DSM Axis I

Disorders) (Spitzer, Williams, Gibbons, & First, 1996) and Axis II disorders (e.g.,

the Structured Clinical Interview for DSM Axis II Disorders) (First, Gibbon,

Spitzer, Williams, & Benjamin, 1997). There are also self-report questionnaires to

assess for specific symptoms such as the Beck Depression Inventory to assess for

depression symptoms (Beck, Steer, & Brown, 1996). Given the breadth of the area,

we focus solely on anxiety disorder-specific tools.

22

2 Considerations for Assessment