Review
Utilizing placebo mechanisms for dose reduction in pharmacotherapy

https://doi.org/10.1016/j.tips.2011.12.001Get rights and content

The knowledge and systematic application of the placebo effect remains limited, although its importance to the treatment of various medical conditions has increasingly been recognized. A possible application of the placebo effect to pharmacotherapy is seen in conditioning processes that aim at a placebo-controlled dose reduction of drugs while maintaining the efficacy of the medical treatment. The pairing of a placebo and a pharmacological agent may achieve satisfactory treatment outcomes in combination with a lower dose of medication. This procedure includes classic and instrumental conditioning processes that involve both conscious and non-conscious information processing. Although recent studies have gathered preliminary evidence for the efficacy of placebo-controlled dose reduction (e.g. in psoriasis and attention deficit hyperactivity disorder [ADHD]), they have also illustrated the difficulties that are inherent to this approach. We critically review previous approaches and discuss designs for clinical trials that seem appropriate to the investigation of conditioned placebo effects in pharmacotherapy.

Section snippets

Placebo responses in pharmacotherapy

Placebo-controlled dose reduction (PCDR) describes a procedure by which a certain amount of a pharmacological treatment is replaced by an ‘inert’ placebo while maintaining the efficacy of the treatment. PCDR capitalizes on classic conditioning mechanisms and differential reinforcement rates of the medication. The reinforcement rate is gradually decreased (i.e. the pharmacological agent is intermittently replaced by a placebo). This may happen with or without the patient's knowledge. Therefore,

‘Pavlovian pharmacology’

Classic conditioning represents a form of associative learning. The starting point is an unconditioned stimulus (UCS; e.g. pharmacological properties of a beta-blocker) that naturally leads to an unconditioned response (UCR; e.g. blood pressure reduction). By pairing the UCS with a neutral stimulus in the acquisition phase (e.g. application by capsule), the neutral stimulus becomes a conditioned stimulus (CS). In the evocation phase, it then elicits a conditioned response (CR) that approximates

Instrumental conditioning

From a mechanistic viewpoint, instrumental conditioning involves a behavior that is reinforced according to its occurrence and thus shapes a response–outcome association. The frequency of the respective behaviour will change according to the nature of the reinforcement. Cognitive views have challenged this perspective, because instrumental conditioning is often mediated by conscious expectations of the outcome 16, 17. In pharmacological conditioning, instrumental conditioning occurs when an

Evidence of conditioning in pharmacotherapy

Various studies have established the feasibility of pharmacological conditioning in animal models. In healthy humans, proof-of-principle studies have demonstrated that physiological responses can be conditioned to placebos. This review focuses on placebo effects in the treatment of existing medical conditions in humans (i.e. the therapeutic benefit of a systematic integration of placebo effects into clinical practice). The following research demonstrates that placebo effects can be obtained by

Conditioning effects in expected random assignment of placebo and verum

Participants of placebo-controlled clinical trials are typically informed that they will be administered either the placebo or the pharmacological agent based on random assignment. Thus, participants may hope to receive the verum treatment and develop positive outcome expectancies. If conditioning procedures are used in such trials, the conditioning effect cannot be reliably isolated from the expectancy effects.

Early research often lacks a detailed description of the informed consent process 22

Conditioning effects: deceptive information

Sometimes placebo trials employ deceptive information to induce expectancy effects. This deception may be an assertion that the substance is definitely the verum treatment, even though placebos are administered. This procedure is likely to enhance the expectancy effects. Thus, these studies offer the poorest evidence for conditioning effects. For example, patients with PD demonstrate placebo reactions in response to placebo application after preconditioning with levodopa [31] when a positive

Conditioning effects and open-label use of placebo

A pilot study [36] successfully applied classic conditioning to the bronchodilator response in asthmatic children; the presentation of the CS (distinctive odour) without the bronchodilator inhalation induced a placebo response. In a single case study of open-label PCDR in a child with systemic lupus erythematosus [37], a 50% PCDR of cyclophosphamide led to significant clinical improvement. After its validation in a pilot study [38], a PCDR of amphetamine salts in the clinical management of ADHD

Isolating conditioning effects from expectancies

Current research highlights the clinical relevance of placebo mechanisms such as behavioral conditioning and intermittent reinforcement in a broad range of medical conditions. There is good evidence for conditioning effects in immune-related diseases, but evidence of conditioning in other medical conditions is currently limited. Unfortunately, studies seldom allow for an unequivocal dissection of conditioning and expectancy effects in PCDR. However, placebo research in healthy humans provides

Designing trials of conditioning in pharmacotherapy

We next examine the literature to clarify principles of effective study design. Although each medical condition and each medication will necessitate special adjustments, this section highlights general issues that merit consideration.

Conditioned stimulus and acquisition phase

Any stimulus of the conditioning environment may become the CS, such as the treatment room or the doctor's white coat 9, 12, 19, 42. Previous research has studied the properties of a stimulus that will increase its suitability for conditioning [8]. This pertains to the concept of preparedness and belongingness (i.e. certain sensory stimuli are more readily associated with an US than others 43, 44). An example is seen in the preferred conditioning of immune functions to gustatory stimuli [45].

Optimal reinforcement schedules

In both long-term clinical applications of PCDR reviewed, the study designs included instrumental conditioning. One study employed a predetermined reinforcement rate [20] that decreased to a minimum of 25% reinforcement. The other study [39] did not refer explicitly to instrumental conditioning. Yet, in the evocation phase, the placebo was paired with 50% of the drug dose that had been judged as optimal for each patient. This may represent instrumental conditioning, assuming that 50% of the

Informative control groups

In animal models of pharmacological conditioning, the control groups that are necessary for the internal validity of a trial have been outlined elsewhere [53]. Even in the well-validated paradigm of conditioned immunomodulation, as many as five control groups are needed, if novel CSs, UCSs or different components of the immune system are assessed. By contrast, the research we have reviewed has employed simpler designs. An electrophysiology study in PD was restricted to intra-individual

Crucial issues to be considered

The reviewed studies indicate directions for future research, although further refinement of research designs is needed. Many issues of placebo conditioning in pharmacotherapy remain unanswered. However, we can speculate that it may prove of interest mainly for chronic medical conditions that require long-term medication, for treatments with severe side-effects and for expensive treatments. Under these circumstances, patients may benefit most from the systematic use of placebo mechanisms.

Concluding remarks

Classic conditioning with one evocation phase provides the proof-of-principle that a medical condition is amenable to the placebo effect. To achieve more stable CRs, instrumental conditioning (intermittent reinforcement) is employed. Recent trials of PCDR indicate that therapeutic gains can be maintained if placebos are combined with subtherapeutic drug regimens. Thus, PCDR has the potential to reduce the amount of medication needed for the treatment and to alleviate the side effects for the

Competing interests

B.K.D. declares no conflicts of interest that could have influenced the content of the manuscript. W.R. declares no conflicts of interest that could be viewed as inappropriate influence on the content of this manuscript. He has received honoraria for consultations and presentations on placebo mechanisms and adherence from Astra Zeneca, Heel and Berlin Chemie.

Authors’ contributions

Both authors have been involved in drafting the manuscript, critically revised it and contributed important intellectual content. They have given final approval of the version to be published.

Acknowledgments

This work was supported by grants from the German Research Foundation (DFG) (RI 574/21-1, RI 574/22-1).

References (61)

  • H.S. Javitz

    The direct cost of care for psoriasis and psoriatic arthritis in the United States

    J. Am. Acad. Dermatol.

    (2002)
  • M.A. Flaten

    Drug effects: agonistic and antagonistic processes

    Scand. J. Psychol.

    (2009)
  • R. Ader

    Conditioned responses in pharmacotherapy research

    Psychol. Med.

    (1993)
  • W. Rief

    Mechanisms involved in placebo and noceco responses and implications for drug trials

    Clin. Pharmacol. Ther.

    (2011)
  • D.D. Price

    A comprehensive review of the placebo effect: recent advances and current thought

    Ann. Rev. Psychol.

    (2008)
  • F. Benedetti

    How placebos change the patient's brain

    Neuropsychopharmacology

    (2010)
  • L. Colloca et al.

    Role of expectations in health

    Curr. Opin. Psychiatr.

    (2011)
  • S. Stewart-Williams et al.

    The placebo effect: dissolving the expectancy versus conditioning debate

    Psychol. Bull.

    (2004)
  • F. Benedetti

    Conscious expectation and unconscious conditioning in analgesic, motor, and hormonal placebo/nocebo responses

    J. Neurosci.

    (2003)
  • D.S. Ramsay et al.

    Biological consequences of drug administration: implications for acute and chronic tolerance

    Psychol. Rev.

    (1997)
  • S. Siegel

    Pavlovian conditioning: the associative basis of tolerance

    Exp. Clin. Psychopharm.

    (2000)
  • F. Benedetti

    Mechanisms of placebo and placebo-related effects across diseases and treatments

    Annu. Rev. Pharmacol. Toxicol.

    (2008)
  • R. Eikelboom et al.

    Conditioning of drug-induced physiological responses

    Psychol. Rev.

    (1982)
  • I. Kirsch

    The role of cognition in classical and operant conditioning

    J. Clin. Psychol.

    (2004)
  • D.R. Shanks

    Learning: from association to cognition

    Annu. Rev. Psychol.

    (2010)
  • F. Benedetti

    Inducing placebo respiratory depressant responses in humans via opioid receptors

    Eur. J. Neurosci.

    (1999)
  • R. Ader

    The role of conditioning in pharmacotherapy

  • R. Ader

    Conditioned pharmacotherapeutic effects: a preliminary study

    Psychosom. Med.

    (2010)
  • C.L. Zimmer-Hart et al.

    Extinction of Pavlovian conditioned inhibition

    J. Comp. Physiol. Psychol.

    (1974)
  • L.M. Greenberg et al.

    Differential effect of abrupt versus gradual withdrawal of chlorpromazine in hospitalized chronic schizophrenic patients

    Am. J. Psychiatry

    (1966)
  • Cited by (102)

    • Human psychoneuroimmunology

      2023, Encyclopedia of Mental Health, Third Edition: Volume 1-3
    View all citing articles on Scopus
    View full text