The lack of scientific attention devoted to the placebo effect as a phenomenon in its own right probably reflects the paucity of theoretical positions within which to organize the existing data and design new research. The proposed investigation 1) is an attempt to advance from a descriptive to an experimental analysis of the placebo effect, taking into account classical conditioning effects, and 2) examines the clinical implications of partial reinforcement as it is applied to the treatment of insomnia. Subjects with primary insomnia will be treated with zolpidem for a period of one month and then randomized to one of four groups for a period of 12 weeks: one receiving full dose zolpidem on a nightly basis (continuous reinforcement), one receiving full dose zolpidem on 14 of 28 nights where placebo is provided on non-drug nights (partial reinforcement), one receiving full dose zolpidem on 14 of 28 nights where no pills are imbibed on non-drug nights (intermittent dosing), and one receiving 5 mg dose zolpidem on a nightly basis (continuous reinforcement with half the standard dose). Following treatment, subjects will be entered into an extinction protocol during which they will 1) continue on the schedule assigned during the experimental period, 2) receive only placebo, or 3) receive neither drug nor placebo. Sleep and daily functioning will be monitored on a daily basis via sleep diaries for the duration of the study. It is hypothesized that, holding cumulative dose constant, a partial schedule of reinforcement will enable patients to better maintain their clinical gains as compared to subjects that receive either continuous reinforcement with half the standard dose or half the frequency of use.

Relevance: The proposed research is not an attempt to offer a behavioral alternative to drug treatment; it is an attempt to acknowledge and capitalize on a behavioral dimension in the design of drug treatment protocols. The value of the proposed research resides in its capacity to provide for the long term treatment of insomnia in a manner that increases the durability of pharmacotherapy while reducing the overall amount of medication required. If proven effective in the current application, this new approach to pharmacotherapy and placebo effects is likely to stimulate new interdisciplinary research for the treatment of a variety of chronic diseases.

Inclusion Criteria:

• Patients with insomnia will meet RDC criteria for psychophysiologic insomnia(99). These criteria are provided in Appendix 2. In addition, the complaint of disturbed sleep will have one or more of the following characteristics:

  • > 30 minutes to fall asleep (Initial Insomnia)
  • 2 awakenings per night of >15 minutes duration and/or wake after sleep onset time of > 30 minutes (Middle Insomnia)
  • An awakening of > 30 minutes prior to the desired "wake up" time (Late Insomnia)
  • Any two of the above complaints (Mixed Insomnia)

Additionally, total sleep time will not exceed 6 hours (unless the sleep efficiency quotient is < 80%) and the problem frequency must be equal to or greater than 4 nights/ week (severe insomnia) with a problem duration > 6 months (chronic insomnia). This profile must be evident at both intake (based on retrospective reports) and as an average profile from the two weeks of baseline diaries (based on prospective sampling).

Exclusion Criteria:

• Unstable medical or psychiatric illness Assessed with the Mini International Neuropsychiatric Interview (MINI) and the The Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADS-L) To assure that the insomnia is not secondary to these factors
• Symptoms suggestive of sleep disorders other than insomnia Assessed with the SDS-CL To assure that the insomnia is not secondary to these factors
• Polysomnographic data indicating sleep disorders other than insomnia Assessed with PSG in collaboration with our sleep medicine consultants To assure that the insomnia is not secondary to these factors
• History of head injury with a sustained loss of consciousness Assessed by self report during the Intake Interview To help assure that the EEG measures are unconfounded by brain damage
• Evidence of active illicit substance use or fitting criteria for alcohol abuse or dependence Assessed with a structured psychiatric interview schedule (the MINI) , written versions of clinical interview queries regarding alcohol use, abuse and dependence (the AUDIT and CAGE), the toxicology screen which is part of the clinical chemistries obtained during the screening physical. To assure that the insomnia is not secondary to these factors and to assure that substance use/abuse does not confound treatment.
• Use of CNS active medications, antidepressants, and hypnotics other than zolpidem Assessed by self report and from the toxicology screen which is part of the clinical chemistries obtained during the screening physical. To help assure that the clinical effects observed in this study are due to the study medication and schedule of reinforcement.
• Inadequate language comprehension Informally, assessed by the Clinical Research Coordinator during Intake Interview To assure the quality of self report data as all the measures are in English.
• Pregnancy Assessed by self report and from the clinical chemistries data obtained during the screening physical. Excluded so as to 1) prevent the fetus from exposure to the study medication (although it should be noted that the medication is considered FDA pregnancy category B) and 2) control for the biopsychosocial changes that occur with pregnancy and may alter the response to the study medication and schedule of reinforcement.
• No first-degree relatives with bipolar disorder or schizophrenia Assessed by self report and a structured psychiatric interview schedule (the SADs). Excluded to reduce risk for first onset during the study

Analyses were only conducted among participants who were compliant with their medication regimens during Phase 3.