The Natural History of Insomnia: What We Know, Don’t Know, and Need to Know

Article information

Sleep Med Res. 2011;2(3):79-88
Publication date (electronic) : 2011 December 31
doi :
1Behavioral Sleep Medicine Program, Department of Psychiatry, University of Pennsylvania, Philadelphia, PA, USA
2The Northumbria Centre for Sleep Research, Department of Psychology, Northumbria University, Newcastle, UK
Correspondence: Michael Perlis, PhD, Behavioral Sleep Medicine Program, Department of Psychiatry, University of Pennsylvania, Suite 670, 3535 Market Street, Philadelphia, PA 19104, USA, Tel +1-215-746-3577, Fax +1-270-512-9829, E-mail
Received 2011 December 06; Accepted 2011 December 13.


To date, studies of the natural history of insomnia have focused on the prevalence, incidence, and persistence of chronic insomnia. While these studies have provided seminal information about the epidemiology of insomnia, no studies to date, have been conducted in a manner to 1) allow for a close resolution of the “transitions” from good sleep to acute insomnia, from acute insomnia to the recovery of good sleep, or from acute insomnia to chronic insomnia and/or 2) allow for a comprehensive assessment of the factors that have been theorized to mediate or moderate these transitions.14 The present paper provides a review of these issues and sets forth a research agenda.

What We Know and Don’t Know (Empirical Findings)

Over the last several decades the prevalence and incidence of insomnia has been described in great detail. These data, while essential to document the magnitude of this health problem, infrequently provide information related to the incidence of new-onset insomnia and even less frequently provide information on the incidence of spontaneous remission and relapse. Of the groundbreaking studies that document these phenomena, none have provided information about the factors that mediate/moderate the transitions between good sleep to acute insomnia and from acute insomnia to either recovery or chronic insomnia. In this review, what is known about the natural history of insomnia (in terms of prevalence, incidence, and clinical course) will be reviewed. In addition, the leading theoretical perspectives on insomnia will be reviewed (with an eye towards identifying the factors that may mediate/moderate the above noted transitions) and a research agenda provided.


The prevalence of insomnia has been examined in several dozen studies. These studies were comprehensively reviewed by Ohayon5 in 2002. One of the most important findings from this literature was that the prevalence varied considerably depending on how insomnia was defined (Fig. 1). Many studies used one or more questions pertaining to symptoms of insomnia, asking respondents to indicate whether they had trouble falling asleep, staying asleep, or with early morning awakenings. When no frequency/severity/chronicity criteria were specified, prevalence rates ranged from 30% to 48% of respondents. When frequency considerations were taken into account, as well as the complaint of daytime consequences, then the prevalence rates were reduced to between 6% and 18%. The latter definitions, while not taking into account severity or chronicity, do approximate the level of detail required for formal diagnosis using current nosologies6,7 and thus likely represent the prevalence of insomnia as a disorder in the population at large. While it is important to estimate point prevalence (to define how widespread a disorder or disease is), such data do not address several related issues: 1) what is the relative prevalence of acute and chronic insomnia; 2) how frequently do acute episodes of insomnia occur within the individual; 3) how many episodes of acute insomnia occur, on average, before the insomnia condition becomes chronic; and 4) what factors mediate/moderate the transition from acute to chronic insomnia.

Fig. 1.

Insomnia symptoms pyramid.

To our knowledge, only one study has specifically assessed lifetime prevalence. Breslau et al.8 used the Diagnostic Interview Schedule to assess lifetime prevalence of DSM-III-R diagnoses in a sample of 21–30 year-olds. The lifetime prevalence of insomnia, defined as ≥ 2 weeks of trouble falling or staying asleep, was 24.6%. In an older sample, the lifetime prevalence would likely be substantially higher if for no other reason than increased opportunity to experience insomnia, not to mention the cumulative effects of aging and illness. Perhaps a more reasonable estimate can be drawn from a recent study on the long-term course of insomnia reported by Buysse et al.9 In this 20-year prospective study 591 adults completed six separate interviews over the course of two decades. Sleep questions from each interview were used to determine both the occurrence of insomnia symptoms as well as a insomnia diagnosis using a proxy definition. Approximately 70% in this cohort reported insomnia symptoms during at least one interview, indicating a very high 20-year prevalence. These data, while suggestive, do not allow for a determination of 1) the lifetime prevalence of acute and chronic insomnia; 2) how frequently acute insomnia occurs within the individual over the lifespan; 3) the factors mediate/moderate the transition from acute to chronic insomnia.

Incidence of Insomnia

Surprisingly, there are eleven studies that provide data regarding rates of new-onset insomnia (variously defined) in individuals who were initially assessed as good sleepers.8,10,1119 This is surprising because such data are, more often than not, contained within investigations that have as primary end points, the assessment of prevalence. Thus, the incidence data may not be featured in the titles and/or abstracts of such reports. Follow-up periods for the 11 studies ranged from 1 to 12 years. The average per-year incidence rate across studies was 5% with a range of 1% to 15% (Fig. 2). The lowest rate was found in a sample consisting only of men.15

Fig. 2.

One year incedence rate of insomnia.

The highest rates were found in a sample of people with chronic medical conditions,13 and in a sample taken from general practitioner offices.16 These findings are not surprising given that medical illness, especially those involving physical pain, can act as a precipitant of acute insomnia. In the studies that examined predictors of new-onset insomnia, risk factors included female gender, depression, poor health, and low physical activity.11,12,14 It should be noted that, contrary to what one might imagine, studies of older adult samples did not find higher incidence rates than younger samples.12,14 As with prevalence studies, insomnia definitions affected study results. For example, Katz and McHorney.13 in a study of individuals with chronic medical and psychiatric conditions, computed separate incidence rates for ‘severe’ and ‘mild’ insomnia, a distinction not made in other studies, and found widely disparate incident rates of 3% and 21%, respectively. This finding, while underlining the importance of how insomnia is defined, is nevertheless consistent with what one might expect, which is that a broader definition of insomnia will yield higher incidence rates. To date, no studies have yet determined: 1) the incidence rate for insomnia using quantitative criteria (along with traditional diagnostic criteria) or narrow sampling intervals (1–3 month repeated assessments), and 2) the relevance of factors that have been hypothesized to increase risk for new onset acute and chronic insomnia.

Persistence of Insomnia Over Time

A total of 25 studies of insomnia have been conducted that included a longitudinal component assessing chronicity.811,1332 The majority were epidemiological studies that included samples of both good and poor sleepers. Follow-up periods ranged from 4 months to 20 years. The findings from these studies, which in part constituted the source data for the Ohayon meta-analysis,5 are represented in Fig. 3. The mean rate of persistence (insomnia at baseline and at follow-up) was 54.8% with a range of 13% to 88%. The highest rates were in longitudinal followups of patients from sleep clinics27,31 and are hence less representative of the overall population because their samples consisted of solely treatment-seeking individuals. Persistence rates were higher when only individuals with severe insomnia were examined.13 As with incidence, greater persistence was associated with chronic medical and psychiatric conditions.14,32 One study explored several variables, based on constructs related to the etiology of insomnia, and found greater persistence to be associated with greater reported somatic arousal and more highly endorsed dysfunctional beliefs and attitudes about sleep.32 Interestingly, longer follow-up periods were not associated with lower rates of persistence (i.e., it seems to follow that the longer follow-up period, the greater the opportunity for recovery-and this was not the case).

Fig. 3.

Persistence of insomnia.

To date, the most comprehensive natural history of insomnia study has been undertaken by Morin et al.21 They conducted a longitudinal study of 388 adults with insomnia in which there were annual assessments each year for three years. One of the many strengths of this study is that subjects were operationally defined as good sleepers or having insomnia based on current research diagnostic criteria, rather than relying on non-standard assessments and definitions of insomnia. One of the seminal findings of this study (to date) is that insomnia was shown to persist over time, with 74% of subjects reporting insomnia for at least one year. Even in those who experienced remission of symptoms at one assessment, 27% eventually had a relapse. Greater severity of insomnia at baseline was associated with higher rates of persistence over time, as one might expect.

The data from these studies suggest that once insomnia persists for some unknown time period, remission is unlikely. The fundamental question that remains to be empirically determined regarding the persistence of insomnia is: “How persistent must the insomnia be before it is truly self-perpetuating and chronic”.

Synopsis of Empirical Findings

The combination of the studies summarized above suggest that insomnia is highly prevalent (with rates up to 48%), occurs with an annual incidence rate of approximately 5%, and tends to be unremitting for 55% of affected individuals. One of the major limitations of the longitudinal studies thus far is that they provide only a snapshot of the course of insomnia that is low in resolution, with most assessments occurring with a frequency of once per year. A finer grained analysis is needed in order to resolve phenomena that are conceptualized as occurring within shorter time frames. For example, it’s not possible to characterize the transition from acute to chronic insomnia with assessments once per year; this transition is thought to occur in time frames from one to three months.6,7,33 These time frames, it should be noted, have more to do with traditional conceptualizations of what constitutes chronic than an empirical assessment of how severe and how persistent insomnia needs to be before it is improbable that it will spontaneously resolve (i.e. exist in a chronic form). As important, the current studies only provide descriptive, numerical analyses regarding prevalence, incidence, and persistence. They do not provide information regarding the process of how insomnia develops over time. For example, while it may be the case that 45% of subjects with acute insomnia experience remission or recovery, very little is known about what differentiates these individuals from those who develop chronic insomnia. In order to accomplish this, longitudinal studies would need to be deployed that have a high sampling resolutions (e.g., once every 1–12 weeks) and that directly measure the factors that have been posited to account for how insomnia becomes chronic. Below the leading theoretical perspectives on insomnia (Human Models) will be reviewed with an eye towards identifying the factors that may mediate/moderate the transitions from 1) good sleep to acute insomnia, 2) acute insomnia to the recovery of good sleep, and 3) acute insomnia to chronic insomnia.

What We Know (Models of Insomnia)

Up until the mid-1990s there were very few theories regarding the etiology and pathophysiology of insomnia and only two gained widespread acceptance and served as platforms for treatment: the Stimulus Control Model34,35 and the 3 Factor Model.1,36 In more recent years, there has been a focus on how neurophysiologic and cognitive processes may contribute to the incidence, severity and chronicity of insomnia. These perspectives (Neurocognitive, Morin, Harvey and Psychobiological Inhibition Models) are briefly reviewed below.

Stimulus Control Perspective (1972)

Stimulus control, which is based on principles of instrumental conditioning, was originally applied to the problem of insomnia by Bootzin and Nicassio.37 The general principle is that one stimulus may elicit a variety of responses, depending on the conditioning history. When a stimulus is always paired with a single behavior there is a high probability that the stimulus will yield only one response. In a more complex conditioning history, a stimulus is paired with a variety of behaviors and this leads to a low probability that the stimulus will yield only one response. In individuals with insomnia, it is frequently the case that large amounts of time are spent in the bedroom and/or the bed itself engaging in non-sleep related behaviors (reading, watching TV, working, spending time on the internet, worrying, etc.). From the patient point of view, engagement in these behaviors is justified because 1) leaving the bedroom would make it more likely that the bout of wakefulness would be prolonged and 2) staying in bed is restful and thus is somewhat recuperative. From the conditioning point of view, the increase in the number of non-sleep behaviors in the bed and bedroom leads to stimulus dyscontrol: the reduced probability that sleep-related stimuli will occur in association with sleepiness and sleep. Stimulus dyscontrol can therefore be conceptualized as a problem that grows out of a normal process: instrumental conditioning. Further, the problem consists of two components: 1) time spent engaged in non-sleep activities in the bed and bedroom and 2) time spent in bed “awake” (i.e., wakefulness is a non-sleep activity). The latter, while not an explicit focus of the stimulus control perspective, is also thought to be problem that grows out of a normal process: classical conditioning. In this instance, the repeated pairing of bed and bedroom with the physiologic/neurobiologic state of wakefulness makes it likely that these stimuli may become conditioned stimuli, not for sleepiness and sleep (as is normally the case), but for wakefulness itself. Interestingly, patients often allude to this state of affairs when they report being sleepy while watching TV in the living room only to become inexplicably alert and awake upon crossing the threshold into the bedroom.

Most would agree that the stimulus control perspective is compelling. Further, the important role of stimulus dyscontrol as a perpetuating factor for insomnia is supported by the finding that Stimulus Control Therapy (which involves reestablishing a one-to-one correspondence between sleep-related stimuli and sleep by eliminating non-sleep activities in the bed and bedroom) is highly efficacious.38,39 This said, the efficacy of the therapy (while important clinically) is not proof positive that stimulus dyscontrol is, in part or large measure, pathogenic. To date, no prospective study has been undertaken to shown that the transition from acute insomnia to chronic insomnia is associated with an increased occurrence of (or time spent in) non-sleep related behaviors in the bed and bedroom. Further, there has been no direct demonstration that the bed/bedroom acts as a conditioned stimulus for wakefulness, increased alertness, or “hyperarousal” (in e.g., chronic but not acute insomnia), although there is one self report study that supports this concept.40

The Three Factor Model (1987)

This model (also referred to as the Spielman model, the Behavioral model, or the “Three P model”) is essentially a stress-diathesis model with an additional component that delineates how acute insomnia becomes chronic.1 In brief, this model posits that insomnia occurs acutely in relation to both trait (predisposing factors) and life stress (precipitating factors) considerations. Chronic insomnia occurs as a result of maladaptive coping behaviors (perpetuating factors). Predisposing factors include: trait hyperarousal; personality characteristics such as the tendency to worry or ruminate; and social factors, such as incompatible sleep schedules between bed partners and social pressures to sleep according to a non-preferred sleep schedule (e.g., child rearing). Precipitating factors are acute occurrences that trigger sleep continuity disturbance. The primary “triggers” are thought to be related to life stress events (e.g., perceived or real threat) and medical and psychiatric illness. Perpetuating factors refer to the behaviors adopted by the individual, which are intended to compensate for sleep loss. In the original conceptualization of the behavioral model, Spielman focused on the role of sleep extension. Extending sleep opportunity refers to the tendency of patients to compensate for sleep loss by napping and/or going to bed earlier (advance of time to bed) and/or by getting out of bed later (delay of time out of bed). Such changes are enacted in order to “recover what has been lost” and thus 1) appears to the patient to be a reasonable strategy for coping with insomnia and 2) is reinforcing to the extent that additional sleep diminishes insomnia related symptoms (fatigue, concentration and memory problems, malaise, etc.). The consequence of sleep extension, however, is the dysregulation of sleep homeostasis. That is, the reduction in the amount of time spent continuously awake can be expected to result in reduced “sleep pressure” and that this in turn serves to perpetuate insomnia in the relative, or complete, absence of the original precipitants.

There is no doubt that the Spielman model is powerful and persuasive. It has high face validity, is based on a well established concept (stress-diathesis), and its “third factor” (sleep extension) is firmly grounded on behavioral principles (instrumental conditioning) and is clearly consistent with the two-process model of sleep wake regulation (i.e., de-priming the sleep homeostat* while keeping sleep opportunity constant, will necessarily result in sleep continuity disturbance). Further, the important role of sleep extension is supported by the findings 1) of low sleep efficiency in patients with insomnia and 2) that Sleep Restriction Therapy, which involves limiting time spent in bed both at night and during the day, is highly efficacious. These lines of evidence are not, once again, “proof positive” of the importance of sleep extension as a primary perpetuating factor for insomnia. To date no prospective study has shown that patients with acute insomnia who develop chronic insomnia do indeed compensate for sleep loss with sleep extension and that this differentiates them from both good sleepers and subjects who recover from acute insomnia. Moreover, there is no evidence that one form of the three possible forms of sleep extension (advancing time to bed, delaying time out of bed, or napping) is more damaging than the other two and/or that the various forms sleep extension when combined have additive or multiplicative effects with respect to insomnia morbidity.

The Neurocognitive Model (1997)

The Neurocognitive model37,41 is based on, and is an extension of, the Spielman model.42 The central tenets of the model include

  1. A pluralistic perspective of “hyperarousal”. That is hyperarousal may exist within several domains (alone or in combination) including the cortical, cognitive and somatic domains.

  2. The specification that cortical arousal is central to the etiology and pathophysiology of insomnia (as opposed to cognitive or somatic arousal).

  3. The proposition that cortical arousal, in the context of chronic insomnia, occurs as a result of classical conditioning and is permissive of processes that do not occur with normal sleep (i.e., increased sensory and information processing at sleep onset and during NREM sleep).

  4. The proposition that sleep initiation and maintenance problems do not occur because of “hyperarousal” per se, but because of increased sensory and information processing (i.e., the increased likelihood of detecting internal or external events and experiencing startle and/or orienting responses to such events).

  5. The suggestion that sleep state misperception derives from an attenuation of the normal mesograde amnesia of sleep (i.e., the recollection of events that occur during the sleep period serves to blur the phenomenologic distinction between sleep and wakefulness).

  6. The hypothesis that conditioned cortical arousal is self reinforcing. That is, if sleep related stimuli (X) elicit cortical arousal (Y), and the occurrence of Y reinforces the association of X and Y, then pairing is self-reinforcing. This is particularly important in that this virtually guarantees that the insomnia will 1) in the absence of its original precipitants, continue unabated, and 2) not be subject to extinction as it usually occurs with classical conditioning.

The Neurocognitive model is supported by a series of small scale studies that show: 1) patients with chronic insomnia (as compared to good sleepers) exhibit increased CNS activation, as assessed with quantitative EEG and imaging studies4346; 2) that the regions identified as activated via imaging are associated sensory processing, information processing, and/or long term memory formation46,47; 3) patients with chronic insomnia (as compared to good sleepers) exhibit increased sensory and information process (and also a lack of inhibition of these processes) at around sleep onset and during NREM sleep, as assessed with Evoked Response Potential methodology4345; and 4) patients with chronic insomnia (as compared to good sleepers) exhibit increased long term memory for acoustic stimuli presented at around sleep onset and during NREM sleep.48 To date no prospective study has shown that patients with acute insomnia who develop chronic insomnia do indeed exhibit alterations in sensory processing, information processing, and/or long term memory formation and that this differentiates them from subjects who recover from acute insomnia.

Cognitive Models

The cognitive perspective has been, in principle, around since time immemorial and is essentially the proposition that worry has a central etiologic role in the precipitation and perpetuation of insomnia.49 With respect to the precipitation of insomnia, the concept is that sleep is characterized by, initiated by, and maintained by, the absence of cognitive processes. Therefore, any event or disease process that gives rise to intrusive thoughts, rumination, or perseveration (regardless of the content of the thoughts themselves) should prohibit the initiation and/or maintenance of sleep. Consistent with this view, are the results from a variety of studies that show that patients with chronic insomnia exhibit higher scores on instruments measuring factors related to worry50,51 and that patients with insomnia engage in more sleep-related worry during the pre-sleep period.52,53 Moreover, there is one study, although cross-sectional, that does show differences in worry between normal sleepers, people with acute insomnia and people with chronic insomnia.54 What has not been demonstrated is that intrusive thoughts, rumination, or perseveration (in and of themselves) mediate the transition from good sleep to acute insomnia.

With respect to the perpetuation of insomnia, the role of worry was not formally explicated until Morin55 suggested that dysfunctional beliefs about sleep (specifically this type of worry) may serve to perpetuate insomnia. This point of view is buttressed by several studies which show that patients with insomnia endorse sleep related dysfunctional beliefs to a greater extent than good sleepers, and that treatment of insomnia is associated with a reduction in these beliefs.56 While most would agree that dysfunctional beliefs play an important role in the perpetuation of chronic insomnia, the empirical findings to date are limited. A strong “proof of concept” is required, one where it is shown that patients with acute insomnia who develop chronic insomnia do indeed exhibit more dysfunctional beliefs and that this differentiates them from subjects who recover from acute insomnia.

More recently, the cognitive perspective has been greatly elaborated to include an information processing perspective. That is, insomnia not only occurs in association with worry but with a fundamental alteration in the way that individuals perceive, remember, and respond to events related to the experience of insomnia. This paradigmatic shift was nearly simultaneously initiated by the group at Oxford University led by Dr. Allison Harvey (now at UC Berkeley) and the group from the University of Glasgow led by Dr. Colin Espie.

According to the Harvey2 model (2002), acute insomnia occurs in association with life stress, sub-chronic insomnia occurs with worry about sleeplessness, and chronic insomnia is maintained by selective attention, distorted perceptions of daytime deficits, and counterproductive safety behaviors. Selective attention refers to the development of an attentional set where the individual regularly monitors the internal (e.g., mental alertness or body sensations) and external (e.g., the bedroom clock or noise) environments for sleep-related threats. This change in attentional set is thought to be automatic and the increased effort to detect sleep related ‘threats’ is presumed to result in increased detection of sleep-related threats. Detection of sleep-related threats increases both cognitive and physiologic arousal and reinforces monitoring behavior and is therefore self-perpetuating. In addition to the sleep-related alterations in attentional processes, Harvey also points to increased attention to the daytime consequences of insomnia (e.g., increased fatigue, irritability and dysphoric mood, gastrointestinal distress, loss of perceived attractiveness, and memory and concentration problems). Individuals with insomnia are concerned about the consequences of sleeplessness and are thought to selectively attend to daytime problems and attribute these problems exclusively to poor sleep. As with monitoring for “sleep-related threats”, monitoring for daytime deficits also increases the chance of detecting both the occasional relevant symptom and/or random symptoms. Unlike the monitoring that occurs at night (which directly interferes with sleep initiation and maintenance), the detection of daytime consequences prompts the engagement of safety behaviors. This term, which is borrowed from the anxiety disorders literature, refers to the engagement of compensatory behaviors which are thought to mitigate the social or personal costs of sleeplessness. For example, if one is particularly irritable owing to sleep loss, one might avoid social interactions and thereby the occurrence of adverse events. From the patient point of view, this may seem a reasonable strategy and a successful one. From the cognitive point of view, the need for “avoidance” is never actually put to the test. The consequence of the strategy is that safety behaviors become self reinforcing and thus ensures that the insomnia always has daytime effects (and this contributes to illness chronicity, if not severity). It should be noted that the behavioral phenomenon of sleep extension could easily be considered one kind of safety behavioral (i.e., a behavioral adaptation to perceived sleeplessness). To our knowledge, Harvey and colleagues do not specify that this is the case.

This important conceptualization is supported by a variety of empirical studies conducted by the Oxford/Berkeley group wherein they have shown that patients with insomnia do indeed appear to excessively monitor their sleep environment,57 selectively attend to sleep related stimuli,58,59 and engage in daytime safety behaviors.60 Whether the cognitive and behavioral factors delineated by this perspective represent primary perpetuating factors for chronic insomnia, versus predisposing factors, remains to be demonstrated. Although there is evidence that children who are ‘at risk’ of developing insomnia demonstrate an attention bias to sleep-related cues.61 Further, the primacy of these factors with respect to the natural history of insomnia have yet to evaluated. That is, whether these factors distinguish between patients with acute and chronic insomnia, and at what point do these factors come into play across the illness trajectory relative to other factors (e.g., instrumental and classical conditioning, cortical activation, etc.).

The Espie cognitive model (2002/2006)3,62 differs from that of Harvey in that it proposes that insomnia occurs in association with 1) the inhibition of sleep related de-arousal (failure to inhibit wakefulness) and 2) attention to the phenomenon of sleep (as opposed to sleep related threats), the intention to “get sleep”, and effort to “achieve sleep”. The model, variably referred to as the “Psychobiological Inhibition Model and/or the A-I-E Model (for attention-intention-effort), posits that acute insomnia occurs because of an alteration in the neurobiologic mechanisms that normally inhibit wakefulness. While the authors do not explicitly address the issue, the alteration may be viewed as inherently pathological or as systemic part of the “flight-fight” response (which depending on the context may or may not be adaptive). The failure to inhibit wakefulness is, in turn, thought to result in two cognitive phenomena. First, when individuals are unable to sleep, attention is drawn to what is otherwise an automatic process. Increased attention prevents the perceptual and behavioral disengagement necessary for sleep onset to occur. Second, attentional focus produces an intention to sleep, where sleep becomes a purposeful goal; one which cannot be attained simply by virtue of the occurrence of attention and intention. Third, effort is expended in achieving “what is intended” and the individual “tries” to fall asleep. This effort, like enhanced attention and intention, serves only to extend wakefulness. Support for the Psychobiological Inhibition Model is provided by eight studies where the findings reliably indicate

  • Sleep-related mental preoccupation may be associated with the transition from acute to chronic insomnia52,63,64

  • That subjects with psychophysiological insomnia exhibit heightened levels of attention bias (as compared to good sleepers and individuals with Delayed Sleep Phase Syndrome subjects)65,66

  • Attention bias to sleep-related stimuli detected in patients with psychophysiological insomnia may be driven by ‘threat’6669

  • That there are positive linear relationships between sleep-related attentional bias and self reported sleep quality and sleepiness65

  • That subjects with psychophysiological insomnia exhibit “effortful preoccupation with sleep”70

Both the Harvey and Espie models provide formidable conceptualizations of the role of cognitive processes in the development of chronic insomnia. The models are similar owing to a common focus on attentional processes. A distinction of the Harvey model is that it is more articulate with respect to the consequences of selective attention being the engagement of ‘safety behaviors’. It is also much more elaborative when it comes to diurnal effects of insomnia and how selective attention and safety behaviors in that context serve to make insomnia self-perpetuating. A distinction of the Espie model is that it is more articulate with respect to how the attentional bias comes to be and how it leads to sleep effort, with the latter proposed as a primary precipitating and perpetuating factor. Finally, the Espie model provides for a paradigmatic shift in how the pathophysiology of insomnia is conceptualized. The central etiologic agent is not hyperarousal, it is the failure to inhibit wakefulness.71 The conceptual distinction may ultimately lead to the proposition that acute insomnia occurs in relation stress induced hyperarousal but that chronic insomnia occurs simply as sustained wakefulness (despite the proper circadian phase for sleep and “normal” homeostatic pressure for sleep).

From a natural history point of view, the two models suggest a clear sequence of events in the transition from good sleep to acute insomnia to chronic insomnia: 1) There is a real or perceived stressor; 2) with a resolution (partial or total) of the stressor there is a attentional shift toward sleeplessness itself and/or toward factors that are thought to interfere with the ability to initiate or maintain sleep; and 3) behavioral changes are enacted to “get more sleep” and/or to fend off the consequences of sleep loss. To date, the sequence as whole has not been evaluated in a longitudinal study (from good sleep to chronic insomnia). Cross sectional studies have provided support for the models to the extent that patients with chronic insomnia exhibit more of these signs than good sleepers, subjects with acute insomnia, or subjects with forms of insomnia that are presumably more biologically based (e.g., delayed sleep phase syndrome). Longitudinal studies with limited sampling and large temporal windows (i.e., two point sampling where the assessments are separated by months) have provided support for the models to the extent that patients with acute insomnia show fewer of the aforementioned phenomena than then when the insomnia is chronic.

What We Need to Know (Towards a Natural History of Insomnia)

One definition for the natural history of a disease is that it “…refers to a description of the uninterrupted progression of the disease in an individual from the moment of exposure to the causal agents until recovery or death”.72 As such, a true understanding of the natural history of insomnia requires 1) agreement about what insomnia is and 2) detailed investigation at the time the insomnia is first developing with a temporal resolution that allows for the detection of behavioral and cognitive changes that over days to weeks. Accordingly, what is needed is a large scale, multi-year, prospective study to evaluate the putative “causal agents” delineated in the above models and how they vary in time, are sequenced, and predict clinical course and illness severity. More specifically, the following issues need to be addressed.

  • 1) The distinction between acute and chronic insomnia needs to be formulated on the basis of data such that an empirically determination is made regarding how frequent the insomnia has to be and/or how persistent the insomnia has to be before it may be considered likely to become chronic (i.e., how much time must pass before it is unlikely that the patient will exhibit a spontaneous remission or recovery).73 For a comprehensive review of this issue, please see Ellis et al.74

  • 2) The assessment and diagnosis of insomnia needs to take into account illness severity (e.g., Sleep Latency of 0–5 minutes = excessively sleepiness, 5–15 minutes = normal, 15–30 = mild insomnia, 30–45 = moderate insomnia, 45–90 minutes = severe, < 90 minutes is suggestive of Paradoxical Insomnia). Such scaling could be done based on historical precedence but would like be more useful if these determinations were made empirically. An ideal approach to this determination would be to both account for population norms [what number has the best fit (sensitivity and specificity) to the occurrence of complaint]73 and how illnesses severity maps onto functional consequences. Put differently, 1) how long does sleep latency or wake after sleep onset need to be considered by the individual as problematic (based on epidemiologic assessment strategies that take into account at least age and sex) and 2) how severe do sleep initiation and maintenance problems need to be to correspond to interpersonal, performance, and/or health deficits. As, or more important, illness severity needs to be tracked over time to directly test the association between chronicity and severity (i.e., is it the case that with time sleep initiation and/or maintenance problems become worse, less severe, or simply stable).

  • 3) While the DSM V will provide quantitative cutoffs for illness frequency and chronicity, it may be argued that these cutoffs should also be re-evaluated using empirical strategies like those specified above. Further, the frequency of insomnia (nights per week) also needs to be tracked over time to directly test this association.

  • 4) The factors that have been hypothesized to predispose, precipitate, and perpetuate insomnia (and to mediate/moderate the transitions from good sleep to acute insomnia, from acute insomnia to the recovery of good sleep, or from acute insomnia to chronic insomnia) need to be directly assessed from within a large scale natural history study (longitudinal study) that has a sufficient “sampling rate” to detect when transitions occur and sufficient power to assess the predictive utility of, alone and in combination, multiple mediators and moderators. Ideal candidates are suggested in Table 1.

Concluding Remarks

In 2005 NIH State of the Science Conference report75 stated that “the paucity of literature describing the natural course of insomnia underscores the need for large-scale longitudinal studies…”. To date there has only been one published study by the Laval group that responded to NIH’s call for additional research. The data from this study have been informative and are likely to continue to yield important findings in the future. This said, the Laval study needs to be followed up with an investigation that has enhanced temporal resolution (ideally daily assessments over months to years) and uses instruments that allow one to assess the constructs of interest (ideally dedicated and validated measures). Such a study will 1) allow for an assessment of the relative importance of the putative causal factors and 2) position the field to develop not only better therapies but also preventive strategies.


Conflicts of Interest

The authors have no financial conflicts of interest.



The term “Sleep Homeostat” is used for its heuristic value, at present no structure has been identified as the “master controller of sleep homeostasis (i.e., a structure that is akin to the SCN).


1. Spielman AJ, Caruso LS, Glovinsky PB. A behavioral perspective on insomnia treatment. Psychiatr Clin North Am 1987;10:541–53.
2. Harvey AG. A cognitive model of insomnia. Behav Res Ther 2002;40:869–93.
3. Espie CA, Broomfield NM, MacMahon KM, Macphee LM, Taylor LM. The attention-intention-effort pathway in the development of psychophysiologic insomnia: a theoretical review. Sleep Med Rev 2006;10:215–45.
4. Richardson GS, Roth T. Future directions in the management of insomnia. J Clin Psychiatry 2001;62(Suppl 10):39–45.
5. Ohayon MM. Epidemiology of insomnia: what we know and what we still need to learn. Sleep Med Rev 2002;6:97–111.
6. American Psychiatric Association. Diagnostic and statistical manual of mental disorders 4th ed (DSM-IV)th ed. Washington DC: American Psychiatric Association; 1994.
7. American Academy of Sleep Medicine. International classification of sleep disorders: Diagnostic and coding manual 2nd edth ed. Westchester, IL: American Academy of Sleep Medicine; 2005.
8. Breslau N, Roth T, Rosenthal L, Andreski P. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 1996;39:411–8.
9. Buysse DJ, Angst J, Gamma A, Ajdacic V, Eich D, Rössler W. Prevalence, course, and comorbidity of insomnia and depression in young adults. Sleep 2008;31:473–80.
10. Ford DE, Kamerow DB. Epidemiologic study of sleep disturbances and psychiatric disorders. An opportunity for prevention? JAMA 1989;262:1479–84.
11. Roberts RE, Shema SJ, Kaplan GA. Prospective data on sleep complaints and associated risk factors in an older cohort. Psychosom Med 1999;61:188–96.
12. Morgan K, Clarke D. Risk factors for late-life insomnia in a representative general practice sample. Br J Gen Pract 1997;47:166–9.
13. Katz DA, McHorney CA. Clinical correlates of insomnia in patients with chronic illness. Arch Intern Med 1998;158:1099–107.
14. Foley DJ, Monjan A, Simonsick EM, Wallace RB, Blazer DG. Incidence and remission of insomnia among elderly adults: an epidemiologic study of 6,800 persons over three years. Sleep 1999;22(Suppl 2):S366–72.
15. Janson C, Lindberg E, Gislason T, Elmasry A, Boman G. Insomnia in men-a 10-year prospective population based study. Sleep 2001;24:425–30.
16. Morphy H, Dunn KM, Lewis M, Boardman HF, Croft PR. Epidemiology of insomnia: a longitudinal study in a UK population. Sleep 2007;30:274–80.
17. Buysse DJ, Finn L, Young T. Onset, remission, persistence, and consistency of insomnia symptoms over 10 years: longitudinal results from the Wisconsin Sleep Cohort Study (WSCS). Sleep 2004;27:A268.
18. Jansson-Fröjmark M, Linton SJ. The course of insomnia over one year: a longitudinal study in the general population in Sweden. Sleep 2008;31:881–6.
19. Dodge R, Cline MG, Quan SF. The natural history of insomnia and its relationship to respiratory symptoms. Arch Intern Med 1995;155:1797–800.
20. Hohagen F, Käppler C, Schramm E, Riemann D, Weyerer S, Berger M. Sleep onset insomnia, sleep maintaining insomnia and insomnia with early morning awakening–temporal stability of subtypes in a longitudinal study on general practice attenders. Sleep 1994;17:551–4.
21. Morin CM, Bélanger L, LeBlanc M, Ivers H, Savard J, Espie CA, et al. The natural history of insomnia: a population-based 3-year longitudinal study. Arch Intern Med 2009;169:447–53.
22. Rodin J, McAvay G, Timko C. A longitudinal study of depressed mood and sleep disturbances in elderly adults. J Gerontol 1988;43:P45–53.
23. Vollrath M, Wicki W, Angst J. The Zurich study. VIII. Insomnia: association with depression, anxiety, somatic syndromes, and course of insomnia. Eur Arch Psychiatry Neurol Sci 1989;239:113–24.
24. Klink ME, Quan SF, Kaltenborn WT, Lebowitz MD. Risk factors associated with complaints of insomnia in a general adult population. Influence of previous complaints of insomnia. Arch Intern Med 1992;152:1634–7.
25. Livingston G, Blizard B, Mann A. Does sleep disturbance predict depression in elderly people? A study in inner London. Br J Gen Pract 1993;43:445–8.
26. Hohagen F, Rink K, Käppler C, Schramm E, Riemann D, Weyerer S, et al. Prevalence and treatment of insomnia in general practice. A longitudinal study. Eur Arch Psychiatry Clin Neurosci 1993;242:329–36.
27. Mendelson WB. Long-term follow-up of chronic insomnia. Sleep 1995;18:698–701.
28. Ganguli M, Reynolds CF, Gilby JE. Prevalence and persistence of sleep complaints in a rural older community sample: the MoVIES project. J Am Geriatr Soc 1996;44:778–84.
29. Hyyppä MT, Kronholm E, Alanen E. Quality of sleep during economic recession in Finland: a longitudinal cohort study. Soc Sci Med 1997;45:731–8.
30. Jansson M, Linton SJ. Psychological mechanisms in the maintenance of insomnia: arousal, distress, and sleep-related beliefs. Behav Res Ther 2007;45:511–21.
31. Rosenthal LD, Dolan DC, Taylor DJ, Grieser E. Long-term follow-up of patients with insomnia. Proc (Bayl Univ Med Cent) 2008;21:264–5.
32. Jansson-Fröjmark M, Lundquist D, Lundquist N, Linton SJ. How is persistent insomnia maintained? A prospective study on 50–60 years old adults in the general population. Br J Health Psychol 2008;13(Pt 1):121–33.
33. Edinger JD, Bonnet MH, Bootzin RR, Doghramji K, Dorsey CM, Espie CA, et al. Derivation of research diagnostic criteria for insomnia: report of an American Academy of Sleep Medicine Work Group. Sleep 2004;27:1567–96.
34. Bootzin DR, Perlis ML. Stimulus Control Therapy. In : Perlis ML, Aloia M, Kuhn B, eds. Behavioral Treatments for Sleep Disorders: A comprehensive Primer of Behavrioal Sleep Medicine Interventions Amsterdam: Academic Press; 2011. p. 21–30.
35. Bootzin RR. Stimulus control treatment for insomnia In : Proceedings of the 80th Annual Convention of the American Psychological Association; 1972. p. 395–6.
36. Spielman AJ, Saskin P, Thorpy MJ. Treatment of chronic insomnia by restriction of time in bed. Sleep 1987;10:45–56.
37. Bootzin RR, Nicassio PM. Behavioral treatments for insomnia. In : Hersen M, Eisler RM, Miller PM, eds. Progress in Behavior Modification 6New York: Academic Press Inc; 1978. p. 1–45.
38. Morin CM, Culbert JP, Schwartz SM. Nonpharmacological interventions for insomnia: a meta-analysis of treatment efficacy. Am J Psychiatry 1994;151:1172–80.
39. Murtagh DR, Greenwood KM. Identifying effective psychological treatments for insomnia: a meta-analysis. J Consult Clin Psychol 1995;63:79–89.
40. Robertson JA, Broomfield NM, Espie CA. Prospective comparison of subjective arousal during the pre-sleep period in primary sleep-onset insomnia and normal sleepers. J Sleep Res 2007;16:230–8.
41. Perlis ML, Giles DE, Mendelson WB, Bootzin RR, Wyatt JK. Psychophysiological insomnia: the behavioural model and a neurocognitive perspective. J Sleep Res 1997;6:179–88.
42. Spielman AJ, Caruso LS, Glovinsky PB. A behavioral perspective on insomnia treatment. Psychiatr Clin North Am 1987;10:541–53.
43. Yang CM, Lo HS. ERP evidence of enhanced excitatory and reduced inhibitory processes of auditory stimuli during sleep in patients with primary insomnia. Sleep 2007;30:585–92.
44. Bastien CH, St-Jean G, Morin CM, Turcotte I, Carrier J. Chronic psychophysiological insomnia: hyperarousal and/or inhibition deficits? An ERPs investigation. Sleep 2008;31:887–98.
45. Kertesz RS, Cote KA. Event-related potentials during the transition to sleep for individuals with sleep-onset insomnia. Behav Sleep Med 2011;9:68–85.
46. Nofzinger EA, Buysse DJ, Germain A, Price JC, Miewald JM, Kupfer DJ. Functional neuroimaging evidence for hyperarousal in insomnia. Am J Psychiatry 2004;161:2126–8.
47. Drummond SP, Smith MT, Orff HJ, Chengazi V, Perlis ML. Functional imaging of the sleeping brain: review of findings and implications for the study of insomnia. Sleep Med Rev 2004;8:227–42.
48. Perlis ML, Smith MT, Orff HJ, Andrews PJ, Giles DE. The mesograde amnesia of sleep may be attenuated in subjects with primary insomnia. Physiol Behav 2001;74:71–6.
49. Borkovec TD. Insomnia. J Consult Clin Psychol 1982;50:880–95.
50. Dorsey CM, Bootzin RR. Subjective and psychophysiologic insomnia: an examination of sleep tendency and personality. Biol Psychiatry 1997;41:209–16.
51. Kales A, Caldwell AB, Preston TA, Healey S, Kales JD. Personality patterns in insomnia. Theoretical implications. Arch Gen Psychiatry 1976;33:1128–34.
52. Harvey AG. Pre-sleep cognitive activity: a comparison of sleep-onset insomniacs and good sleepers. Br J Clin Psychol 2000;39(Pt 3):275–86.
53. Kuisk LA, Bertelson AD, Walsh JK. Presleep cognitive hyperarousal and affect as factors in objective and subjective insomnia. Percept Mot Skills 1989;69(3 Pt 2):1219–25.
54. Ellis J, Cropley M. An examination of thought control strategies employed by acute and chronic insomniacs. Sleep Med 2002;3:393–400.
55. Morin CM. Insomnia: psychological assessment and management New York: Guilford Press; 1993.
56. Carney CE, Edinger JD. Identifying critical beliefs about sleep in primary insomnia. Sleep 2006;29:444–53.
57. Tang NK, Anne Schmidt D, Harvey AG. Sleeping with the enemy: clock monitoring in the maintenance of insomnia. J Behav Ther Exp Psychiatry 2007;38:40–55.
58. Semler CN, Harvey AG. Monitoring for sleep-related threat: a pilot study of the Sleep Associated Monitoring Index (SAMI). Psychosom Med 2004;66:242–50.
59. Semler CN, Harvey AG. An investigation of monitoring for sleep-related threat in primary insomnia. Behav Res Ther 2004;42:1403–20.
60. Harvey AG. Identifying safety behaviors in insomnia. J Nerv Ment Dis 2002;190:16–21.
61. Ellis J, Thomson A, Gregory AM, Sterr A. Biased processing of sleep-related stimuli in children of parents with insomnia. Behav Sleep Med 2011;(In Press).
62. Espie CA. Insomnia: conceptual issues in the development, persistence, and treatment of sleep disorder in adults. Annu Rev Psychol 2002;53:215–43.
63. Taylor LM, Espie CA, White CA. Attentional bias in people with acute versus persistent insomnia secondary to cancer. Behav Sleep Med 2003;1:200–12.
64. Ellis J, Fox P. Promoting mental health in students: is there a role for sleep? J R Soc Promot Health 2004;124:129–33.
65. Jones BT, Macphee LM, Broomfield NM, Jones BC, Espie CA. Sleep-related attentional bias in good, moderate, and poor (primary insomnia) sleepers. J Abnorm Psychol 2005;114:249–58.
66. MacMahon KM, Broomfield NM, Espie CA. Attention bias for sleep-related stimuli in primary insomnia and delayed sleep phase syndrome using the dot-probe task. Sleep 2006;29:1420–7.
67. Marchetti LM, Biello SM, Broomfield NM, Macmahon KM, Espie CA. Who is pre-occupied with sleep? A comparison of attention bias in people with psychophysiological insomnia, delayed sleep phase syndrome and good sleepers using the induced change blindness paradigm. J Sleep Res 2006;15:212–21.
68. Woods H, Marchetti LM, Biello SM, Espie CA. The clock as a focus of selective attention in those with primary insomnia: an experimental study using a modified Posner paradigm. Behav Res Ther 2009;47:231–6.
69. Spiegelhalder K, Espie C, Riemann D. Is sleep-related attentional bias due to sleepiness or sleeplessness. Cognition & Emotion 2009;23:541–50.
70. Broomfield NM, Espie CA. Towards a valid, reliable measure of sleep effort. J Sleep Res 2005;14:401–7.
71. Perlis ML, Pigeon W, Smith MT. Etiology and pathophysiology of insomnia. In : Kryger MH, Roth T, Dement WC, eds. Principles and Practice of Sleep Medicine Philadelphia: W.B. Saunders Co; 2005. p. 714–25.
72. Bhopal RS. Concepts of epidemiology: integrating the ideas, theories, principles and methods of epidemiology 2nd edth ed. New York: Oxford University Press; 2008.
73. Lichstein KL, Durrence HH, Riedel BW, Taylor DJ, Bush AJ. Epidemiology of sleep: age, gender, and ethnicity Mahwah, NJ: Erlbaum; 2002.
74. Ellis JG, Gehrman P, Espie CA, Riemann D, Perlis ML. Acute insomnia: current conceptualizations and future directions. Sleep Med Rev 2012;16:5–14.
75. NIH State of the Science Conference statement on Manifestations and Management of Chronic Insomnia in Adults statement. J Clin Sleep Med 2005;1:412–21.

Article information Continued

Fig. 1.

Insomnia symptoms pyramid.

Fig. 2.

One year incedence rate of insomnia.

Fig. 3.

Persistence of insomnia.

Table 1.

Factors to track

Predispositional factors Precipitational factors Perpetuational factors
Good sleep to acute insomnia Good sleep to acute insomnia Acute to chronic insomnia
 Biological domain   Biological domain   Biological domain
  Sleep need, ability, & plasticity   New onset medical illness4   Chronic medical illness4
  Metabolic rate cumulative   New onset psychiatric illness4   Chronic psychiatric illness4
  Medical burden cumulative   Acute injury4   Conditioned CNS activation5
  Psychiatric burden  Psychological domain   Altered sensory processing6
 Psychological domain   Emergent life stress events  Psychological domain
  Personality factors (N.E.O)1   Cumulative stress   Poor stimulus control7
  Stress reactivity  Social domain   Bedroom as CS for insomnia8
  Current stress   Change in safety of sleep environs   Sleep extension9
  Life stress events2   Change in co-sleeping conditions   Altered information processing10
  Coping style and ability   Change in sleep schedule   Attention bias11
 Social domain   Selective attention & monitoring12
  Safety of sleep environs   Attenuation of normal sleep MA13
  Co-sleeping3   Engagement of safety behaviors14
  New parenthood   Engagement of sleep effort15
 Social domain
  Change in safety of sleep environs
  Change in co-sleeping conditions
  Change in sleep schedule

Neuroticism, introversion/extroversion, openness, etc.


Positive or negative events.


Coherence of preferred sleep phases between partners.


Illness or injury may precipitate (and/or perpetuate) sleep initiation or maintenance problems via pain and/or discomfort… or thru the alteration of sleep need and/or ability.


Neurocognitive model.


Stimulus Control model.


Spielman-3P model.


Psychobiological Inhibition Model, A-I-E Model.


Harvey model.

MA: Mesograde amnesia, CS: conditioned stimulus.