Insomnia and Anxiety (Series in Anxiety and Related Disorders) (27 page)

effect; that is, the unwanted thoughts (and wakefulness) persist even longer (Ree,

Harvey, Blake, Tang, & Shawe-Taylor, 2005). This cycle of unwanted recurring

thoughts has been described in many different disorders as repetitive thinking

(Segerstrom, Tsao, Alden, & Craske, 2000), and the most pertinent types of repetitive

thinking in insomnia are rumination and worry (Carney, Edinger, Meyer, Lindman,

& Istre, 2006; Harvey, 2002; Thoresen, Coates, Kirmil-Gray, & Rosekind, 1981).

The intrusion of this type of thinking has prompted the testing of bedtime

102

7 Sleep-Related Cognitive Processes

arousal-decreasing strategies for processing intrusive material for the day via early

evening problem solving (Carney & Waters, 2006; Espie & Lindsay, 1987) or a

Pennebaker writing (Harvey & Farrell, 2003) assignment. These and other cognitive

strategies will be discussed further in Chap. 8.

The Role of Distress

There have been many studies showing increased emotional arousal in those with

insomnia (Carskadon et al., 1976; Coursey, 1975; Kales, Caldwell, Preston, Healey,

& Kales, 1976; Monroe, 1967). In addition to reporting more distress, people with

insomnia also report taking longer time to emotionally recover from daytime stres-

sors (Waters, Adams, Binks, & Varnado, 1993). One further piece of evidence for

Harvey’s (2002) model is that the presleep thought content of people with insomnia

tends to be negatively valenced (Kuisk et al., 1989). Manipulating presleep distress

(e.g., instructing people that they will have to do a speech upon awakening in the

morning) tends to disrupt sleep (Gross & Borkovec, 1982). Thus, there is support

for distress and emotional arousal in those with insomnia.

Beliefs in Insomnia

Morin (1993) is largely responsible for importing Beck’s Cognitive Theory of

psychopathology (Beck, 1976) into the area of insomnia. In Beck’s classic

Cognitive Theory beliefs are the basis for the automatic thoughts. Early cognitive

conceptualizations focused on unhelpful beliefs as perpetuating and potentially

predisposing factors in insomnia (Morin, 1993). In insomnia, beliefs are thought to

drive sleep-interfering behavior and maintain arousal/distress in the face of poor

sleep. Beliefs also drive a tendency to seek out and pay more careful attention to

information confirming the presumption that poor nighttime sleep will occur or

daytime functioning will be impaired. Research has shown that, in contrast to good

sleepers, people with insomnia have an unhelpful degree of beliefs about sleep that

make them more prone to insomnia (Carney & Edinger, 2006). Collectively, these

beliefs have been shown to respond to CBT for insomnia (Carney & Edinger, 2006;

Edinger, Wohlgemuth, Radtke, Marsh, & Quillian, 2001b; Morin, Blais, & Savard,

2002). These beliefs also show some improvement with relaxation therapy, although

not as much as with CBT (Edinger et al., 2001b). There is less cognitive improve-

ment with pharmacotherapy when compared with CBT for insomnia (Morin et al.,

2002). Presumably, these belief changes relate to decreased helplessness about their

sleep problem (i.e., relaxation therapies and pharmacotherapy produce sleep

improvements). However, despite sleep improvements in all of these treatments,

these beliefs do not change as much as the change associated with a belief-targeted

treatment such as CBT. These CBT belief improvements significantly relate to

other indices of clinical improvement including PSG (Edinger et al., 2001b), sleep

Attention and the “Threat” of Sleep

103

diaries, and other sleep measures such as sleep self-efficacy and a global insomnia

symptom questionnaire (Carney & Edinger, 2006).

Morin (1993) describes two main beliefs in insomnia: (1) that there is something

wrong and there is a sense of lowered self-efficacy about the ability to produce

sleep; and (2) worry about the consequences of poor sleep (Morin, 1993). These

were echoed in Beck’s later writings about beliefs across multiple disorders (Beck,

1999). These two beliefs are mainly concerned with a belief of helplessness to cope

with sleep loss. This is arguably a trans-diagnostic belief that could conceivably

predispose a person to comorbid disorders in addition to insomnia.

In addition to helplessness, a second common theme across such beliefs is the

belief that effort is required to sleep (Espie, Broomfield, MacMahon, Macphee, &

Taylor, 2006). Perhaps, it is the thwarted attempts at sleep effort that lead to the acti-

vation of helplessness related beliefs. Espie and his colleagues used the following

quote from Frankl (1965) to demonstrate this phenomenon: “Sleep (is like) a dove

which has landed near one’s hand and stays there as long as one does not pay any

attention to it; if one attempts to grab it, it quickly flies away.” (Ansfield, Wegner, &

Bowser, 1996) Sleep is something that occurs in the absence of effort. In Espie et al.

(2006) attention–intention effort model, the good sleeper is seen as passive, and sleep

behavior is determined by cues of sleepiness at night and waking cues in the morning.

Such a pattern of reinforcement shapes the pattern without much thought or effort on

the part of the good sleeper. Poor sleep occurs chronically when people begin to pay

attention and exert effort over their sleep. There is evidence of increased sleep-related

effort in those with insomnia relative to good sleepers via self-report (Broomfield &

Espie, 2005) and experimental manipulations that show improved sleep when the

instruction is to stay awake (Ascher & Turner, 1979; Broomfield & Espie, 2003).

Clinically speaking, sleep effort is an important construct because it may be the moti-

vation behind maladaptive sleep behaviors. Going to bed early to
catch-up
on lost

sleep implies that one needs to
do
something to catch-up. Going to bed earlier can

decrease the likelihood for sleep through its deleterious effect on homeostatic sleep

drive. Thus, effort-related beliefs can contribute to important behavioral perpetuating

factors for insomnia, and consequently must be targeted in CBT.

Attention and the “Threat” of Sleep

Do people with insomnia exhibit increased attention to their sleep? One of the

criteria for Psychophysiologic Insomnia in the International Classification of Sleep

Disorders, Diagnostic and Coding Manual, Second Edition (ICSD-2) is an “excessive

focus” on sleep. Studies have supported attentional bias for sleep-related words

(Taylor, Espie, & White, 2003) and sleep-related visual stimuli (Jones, Macphee,

Broomfield, Jones, & Espie, 2005) in people with insomnia relative to normal

sleepers. The importance of attention as a perpetuating factor is not a novel idea and

is in fact invoked in anxiety disorder psychopathology models (Mathews &

MacLeod, 1994). In anxiety disorders, the focus of the heightened attention is on

104

7 Sleep-Related Cognitive Processes

threat-related material (Mogg, Mathews, Bird, & MacGregor-Morris, 1990).

Insomnia is no exception, and the material is presumed to be sleep threat-related

stimuli. Indeed, in the ICSD-2 diagnostic classification scheme mentioned above,

Psychophysiological Insomnia is characterized by heightened anxiety about sleep.

Espie et al. (2006) points to sleep’s prominent place in Maslow’s (1943) Hierarchy

of Human Needs to explain why sleeplessness is so threatening. If sleep is one of

our most basic needs, then it would make sense for considerable resources to be

utilized to remediate this need if sleeplessness were to occur. Several investigations

have also found support for attention focus on sleep-related threat stimuli (Harvey,

2002; Semler & Harvey, 2004; Tang & Harvey, 2004). Sleep-related threat stimuli

includes daytime threats such as scanning the body for fatigue or other symptoms

thought to be associated with poor sleep, as well as nocturnal threats such as scan-

ning the body for symptoms predictive of poor sleep, or focusing on the extended

amount of time it takes to fall asleep (Harvey, 2002).

One example of a sleep-threat stimulus is the bedroom clock. Many people with

insomnia will admit to watching the clock and becoming anxious as they make

calculations of a shortened sleep opportunity. This is demonstrated in a clever

experiment, wherein those with primary insomnia were assigned to one of two

groups: a clock-monitoring group and a digital display unit monitoring group (Tang,

Schmidt, & Harvey, 2006). The digital display unit was identical to the digital clock

in the other condition; however, it was programmed to display random digits that

changed every minute. On the monitoring night, the clock-monitoring group

reported greater sleep-related worry, and longer estimated sleep onset latency (SOL),

relative to baseline and relative to monitoring night data in the digital display unit

monitoring group. The groups did not differ on objective (actiwatch) SOL estimates;

however, in the clock monitoring group, there was a tendency to overestimate sub-

jective SOL, when comparing sleep diaries to objective (actiwatch) SOL estimates.

One implication for this study is support for the role of threat monitoring in

increasing anxiety and potential misperception in sleep estimation. The tendency to

monitor the clock may be indicative of an attentional bias toward threats to sleep.

The perception of a threat to sleep increases anxiety and worry because as time

progresses, less time is available for sleep (Harvey, 2002). Similarly, sleep-related

worry purportedly drives the process of sleep estimation distortion (Borkovec,

1982; Harvey, 2002). Another key implication of this study is support for the

instruction that it may be useful to remove clocks from view as part of insomnia

treatment (Hauri, 1991; Morin, 1993).

The Role of Attributions

There are several studies that support the presumed role of misattribution in cogni-

tive models of insomnia (Harvey, 2002; Harvey, Tang, & Browning, 2005; Morin,

1993). An early study reported that those with insomnia who were told that a placebo

pill would produce arousal symptoms fell asleep faster than insomnia sufferers

Maintaining the Status Quo

105

who were told that the same pill would produce a relaxation response (Storms

& Nisbett, 1970). The explanation? The arousal-pill group attributed their arousal

to the pill rather than to an endogenous (internal) source, thus decreasing anxiety

and decreasing SOL. Or perhaps, when the expected relaxation response was not

detected, it increased arousal, thus increasing SOL. Similarly, another study com-

bined medication with behavior therapy and at the end of the first treatment week,

half of study patients were told that they were receiving a suboptimal dose of the

medication, and the other half were told that they were receiving an adequate dose

(Davison, Tsujimoto, & Glaros, 1973). All participants stopped the drug therapy

and continued with behavior therapy. Those who were told that they had received a

suboptimal dose showed greater maintenance of their improvements than those

who were told that the dose was optimal. Those who were told that the dose was

suboptimal did not attribute their improvement to the drug (because they were told

the dose was not therapeutic). The other group did not maintain their improvements

after stopping the drug because they attributed their improvement to the drug.

Indeed, sleep improvement attributions appear to be important in clinical trials, as

those treated with CBT evidence greater improvements in their confidence in being

able to sleep than those in a control treatment (Edinger, Wohlgemuth, Radtke,

Marsh, & Quillian, 2001a).

Attributions also play an important role in daytime functioning (Morin, 1993).

As part of Harvey’s (2002) Cognitive Model, there is a purported tendency to misat-

tribute daytime symptoms of insomnia, negative mood, or cognitive difficulties to

poor nighttime sleep. Research with the most frequently used measure of maladap-

tive beliefs about sleep, the Dysfunctional Beliefs, and Attitudes about Sleep Scale

(Morin, 1993; Morin, Vallières, & Ivers, 2007) suggests that both primary and

comorbid insomnia groups tend to misattribute daytime symptoms to sleep to a

greater degree than good sleepers (Carney & Edinger, 2006; Morin, 1993).

Misattributing other causes of daytime symptoms such as fatigue increases the

pressure to produce sleep. That is, if feeling poorly during the day is attributed

solely to sleeping poorly (rather than the multitude of possible explanations for

fatigue at any given time during the day), then there will be an increased anxiety

about sleep. Indeed, cognitive restructuring is frequently aimed at correcting such

misattributions (Edinger & Carney, 2008).