Order does matter: the combined effects of classical conditioning and verbal suggestions on placebo hypoalgesia and nocebo hyperalgesia

Conclusions Regarding the Role of Expectations in Placebo Analgesia Studies May Depend on How We Investigate It: A Meta-Analysis, Systematic Review, and Proposal for Methodological Discussions

OBJECTIVE

Expectations are highlighted as a key component in placebo effects. However, there are different approaches to whether and how placebo studies should account for expectations, and the direct contribution has yet to be estimated in meta-analyses. Using different methodological approaches, this meta-analysis and systematic review examines the extent to which expectations contribute to pain in placebo studies.

METHODS

The databases PubMed, PsycINFO, Embase, and Web of Science were searched for placebo analgesia mechanism studies with numerical measures of both expectations and pain. Thirty-one studies, comprising 34 independent study populations (1566 subjects: patients and healthy participants) were included. Two meta-analyses were conducted: meta-analysis 1, using study-level data, estimated the effect of expectation interventions without taking measures of expectations into account (expectations assumed); and meta-analysis 2, using individual-level data, estimated the direct impact of participants' expectations on pain (expectations assessed). Risk of bias was assessed using the Cochrane risk-of-bias tool.

RESULTS

Meta-analysis 1 showed a moderate effect of expectation interventions over no expectation intervention on pain intensity (Hedges g = 0.45, I2 = 54.19). Based on 10 studies providing individual-level data, meta-analysis 2 showed that expectations predicted pain intensity in placebo and control groups ( b = 0.36, SE = 0.05), although inconsistently across study methodologies.

CONCLUSIONS

Participants' expectations contributed moderately to pain in placebo analgesia studies. However, this may largely be influenced by how we measure expectations and how their contribution is conceptualized and analyzed-both within and across studies.

Remifentanil-induced hyperalgesia in healthy volunteers: a systematic review and meta-analysis of randomized controlled trials

ABSTRACT

Recent literature suggests that the withdrawal of remifentanil (RF) infusion can be associated with hyperalgesia in clinical and nonclinical settings. We performed a systematic review and a meta-analysis of randomized controlled trials with cross-over design, to assess the effect of discontinuing RF infusion on pain intensity and areas of hyperalgesia and allodynia in healthy volunteers. Nine studies were included. The intervention treatment consisted in RF infusion that was compared with placebo (saline solution). The primary outcome was pain intensity assessment at 30 ± 15 minutes after RF or placebo discontinuation, assessed by any pain scale and using any quantitative sensory testing. Moreover, postwithdrawal pain scores were compared with baseline scores in each treatment. Secondary outcomes included the areas (% of basal values) of hyperalgesia and allodynia. Subjects during RF treatment reported higher pain scores after discontinuation than during treatment with placebo [standardized mean difference (SMD): 0.50, 95% confidence interval (CI): 0.03-0.97; P = 0.04, I 2 = 71%]. A significant decrease in pain scores, compared with baseline values, was found in the placebo treatment (SMD: -0.87, 95% CI: -1.61 to -0.13; P = 0.02, I 2 = 87%), but not in the RF treatment (SMD: -0.28, 95% CI: -1.18 to 0.62; P = 0.54, I 2 = 91%). The area of hyperalgesia was larger after RF withdrawal (SMD: 0.55; 95% CI: 0.27-0.84; P = 0.001; I 2 = 0%). The area of allodynia did not vary between treatments. These findings suggest that the withdrawal of RF induces a mild but nonclinically relevant degree of hyperalgesia in HVs, likely linked to a reduced pain threshold.

Pain Rating is Worth a Thousand Words: Nocebo Hyperalgesia Induced by Verbal Modeling Prevails Over the Effects of Symbolic Modeling and Verbal Suggestion

This study compares the effectiveness of verbal modeling, symbolic modeling, and verbal suggestion in inducing nocebo hyperalgesia. It is the first study to examine the contribution of stress to observationally induced nocebo hyperalgesia. This study's experimental groups represented various sources of social information: a group of people participating in the study (verbal modeling), a single participant (symbolic modeling), and an experimenter (verbal suggestion). During the experiment, participants received electrocutaneous stimuli at the same intensity, some of which were applied with a nocebo (sham device). Participants in the verbal modeling group were acquainted with pain ratings that had allegedly been provided by other participants. The ratings suggested that other participants experienced more pain in the nocebo trials than in the control trials. In the symbolic modeling group, participants observed a videotaped model experiencing more pain in the nocebo than in the control trials. In the verbal suggestion group, participants received a verbal suggestion of hyperalgesia in the nocebo trials and no suggestion in the control trials. No manipulations were used in the control group. To investigate whether nocebo hyperalgesia is stable over time, an additional extinction phase was conducted. Nocebo hyperalgesia was induced by verbal modeling only and was partially mediated by expectancy. Stress was a significant moderator of the induced effect. Nocebo hyperalgesia was extinguished during the extinction phase. The obtained results provide potential implications for minimizing nocebo hyperalgesia in clinical practice by, for instance, controlling patients' expectancies and stress levels. PERSPECTIVE: The study shows the role of pain-related information derived from other people in shaping negative treatment experiences in the individual. Because information from others has a particular impact on individuals experiencing stress, both this information and the stress level of patients should be monitored in the treatment process.

Learning by observing: a systematic exploration of modulatory factors and the impact of observationally induced placebo and nocebo effects on treatment outcomes

Introduction

Observational learning (OL) refers to learning through observing other people's behavior. OL has been suggested as an effective and simple tool to evoke treatment expectations and corresponding placebo and nocebo effects. However, the exact mechanisms by which OL shapes treatment outcomes, its moderating factors and possible areas of application remain unclear. We thus reviewed the existing literature with two different literature searches to answer the following questions: Which influencing factors contribute to OL-induced placebo and nocebo effects (in healthy volunteers and patients) and how large are these effects (search 1)? In which medical fields has OL been used so far to modulate treatment expectancy and treatment outcomes in patients, their caregivers, and at-risk groups (search 2)? We also aimed to explore whether and how the assessment of treatment expectations has been incorporated.

Methods

We conducted two independent and comprehensive systematic literature searches, both carried out on September 20, 2022.

Results

We identified 21 studies that investigated OL-mediated placebo and nocebo effects for pain and itch, the (placebo) efficacy of sham treatment on anxiety, and the (nocebo) induction of medication side effects (search 1). Studies showed that OL can efficiently induce placebo and nocebo effects across different presentation modes, with medium effect sizes on average: placebo effects, d = 0.79 (range: d = -0.36-1.58), nocebo effects, d = 0.61 (range: d = 0.04-1.5). Although several moderating factors have been investigated, their contribution to OL-induced effects remains unclear because of inconsistent results. Treatment expectation was assessed in only four studies. Regarding medical applications of OL (search 2), we found 12 studies. They showed that OL was effectively applied in preventive, therapeutic and rehabilitative interventions and that it was mainly used in the field of psychosomatics.

Discussion

OL effects on treatment outcomes can be both positive and negative. Future research should investigate which individuals would benefit most from OL and how OL can be implemented most effectively to induce placebo and avoid nocebo effects in clinical settings.

Systematic review registration

This work was preregistered at the Center for Open Science as open-ended registration (doi: 10.17605/OSF.IO/FVHKE). The protocol can be found here: https://archive.org/details/osf-registrations-fvhke-v1.

[Does patient communication influence kinesiophobia? A systematic review]

The aim of the present systematic review was to determine the effect of communication in the health care setting on kinesiophobia. To this end, a literature search was conducted in seven databases between November 2022 and February 2023. The review was carried out following the PRISMA statement and for the analysis of methodological quality we used: PEDro Scale, Van Tulder criteria and risk of bias analysis of the Cochrane Collaboration. A total of 13 articles were included with a mean methodological quality of 7.1 out of 10. Significant results were obtained for at least one variable (kinesiophobia, disability or level of physical activity) in 12 articles. There is strong evidence that communication can influence a subject's kinesiophobia. This influence is most likely to be in a negative or disabling sense, but it can also act in a positive sense by decreasing it.

Creating Placebo Nonresponders in the Lab

Conditioning and expectation are known to be the main mechanisms of placebo analgesia. They may operate together, so that expectations may be enhanced by a conditioning procedure. Although most of the studies have tried to potentiate expectations through conditioning in order to generate good placebo responders, a few studies have tried to mismatch conditioning and expectations in order to investigate the subsequent administration of a placebo. In this study, we want to further investigate this mismatch. We generated incongruent associations during a conditioning procedure in which the study participants did not get what they expected. In fact, although the participants received verbal instructions of pain decrease following the administration of a placebo, we surreptitiously increased the painful electric stimulation. Two pairings of these incongruent associations (mismatch between what was expected and what was experienced) disrupted expectations of analgesia as well as the placebo effect, as assessed by measuring electric pain thresholds on the hand. The effects of mismatch conditioning on the hand extended to the contralateral arm and to a different type of pain (tourniquet), which suggests that local mismatch conditioning may affect the whole body. In all cases, expectations predicted placebo analgesia. These findings indicate that placebo nonresponders can be created in the laboratory by acting on expectations and that local effects can be generalized to other parts of the body. They also stress the importance of expectations in the therapeutic outcome, with important implications for clinical trials. PERSPECTIVE: By using mismatch conditioning, in which study participants did not get what they expected, we reduced expectations of analgesia, and this reduction abolished placebo analgesia. This effect extended to other parts of the body and other types of pain, which indicates that placebo nonresponders can be created in the laboratory.

Does Active Inference Provide a Comprehensive Theory of Placebo Analgesia?

Placebo interventions generate mismatches between expected pain and sensory signals from which pain states are inferred. Because we lack direct access to bodily states, we can only infer whether nociceptive activity indicates tissue damage or results from noise in sensory channels. Predictive processing models propose to make optimal inferences using prior knowledge given noisy sensory data. However, these models do not provide a satisfactory explanation of how pain relief expectations are translated into physiological manifestations of placebo responses. Furthermore, they do not account for individual differences in the ability to endogenously regulate nociceptive activity in predicting placebo analgesia. The brain not only passively integrates prior pain expectations with nociceptive activity to infer pain states (perceptual inference) but also initiates various types of actions to ensure that sensory data are consistent with prior pain expectations (active inference). We argue that depending on whether the brain interprets conflicting sensory data (prediction errors) as a signal to learn from or noise to be attenuated, the brain initiates opposing types of action to facilitate learning from sensory data or, conversely, to enhance the biasing influence of prior pain expectations on pain perception. Furthermore, we discuss the role of stress, anxiety, and unpredictability of pain in influencing the weighting of prior pain expectations and sensory data and how they relate to the individual ability to regulate nociceptive activity (endogenous pain modulation). Finally, we provide suggestions for future studies to test the implications of the active inference model of placebo analgesia.

The relationship between expectancy, anxiety, and the nocebo effect: a systematic review and meta-analysis with recommendations for future research

Despite the theoretical prominence of expectancy and anxiety as potential mechanisms of the nocebo effect, not all studies measure expectancy and/or anxiety, and there are inconsistent findings among those that do. The present study sought to systematically review and meta-analyse available data to evaluate the relationship between expectancy, anxiety and the nocebo effect. The two key questions were: (1) whether nocebo manipulations influence expectancy and anxiety; and (2) whether expectancy and anxiety are associated with the subsequent nocebo effect. Fifty-nine independent studies (n = 3129) were identified via database searches to 1st August 2021. Nocebo manipulations reliably increased negative expectancy with a large effect (g = .837) and state anxiety with a small effect (g = .312). Changes in expectancy and state anxiety due to the nocebo manipulation were associated with larger nocebo effects (r = .376 and .234, respectively). However, there was no significant association between dispositional anxiety and the nocebo effect. These findings support theories that rely on situationally-induced expectancy and anxiety, but not dispositional anxiety, to explain nocebo effects. Importantly, being malleable, these findings suggest that interventions that target maladaptive negative expectancies and state anxiety could be beneficial for reducing the harm nocebo effects cause across health settings. Recommendations for future research are discussed.

Red induces hyperalgesia and white induces hypoalgesia regardless of pain modality

Colors are an important factor that influences different aspects of people's lives. However, little is known about the effects of colors on pain. This preregistered study aimed to investigate whether the type of pain affects the impact of colors on pain intensity. 74 participants were randomly divided into 2 groups according to the type of pain: electrical or thermal. In both groups, pain stimuli of the same intensity were preceded by different colors. Participants rated the pain intensity induced by each pain stimulus. Additionally, pain expectations related to each color were rated at the beginning and the end of the procedure. A significant effect of color on pain intensity ratings was found. Pain was most intense in both groups after red, whereas the lowest ratings were given after white. A similar pattern of results was observed for pain expectations. Expectations also correlated with and were found to be a predictor of experienced pain for white, blue, and green. The study shows that white can reduce, while red can alter the experienced pain. Moreover, it shows that the effect of colors is affected to a greater extent by the pain expectations rather than the pain modality. We conclude that the way colors influence pain broadens the current knowledge on effects of colors on human behavior and could help in the future both patients and practitioners.

Verbal Suggestion Modulates the Sense of Ownership and Heat Pain Threshold During the "Injured" Rubber Hand Illusion

The appearance of the self-body influences the feeling that one's body belongs to oneself, that is, a sense of ownership (SoO) and pain perception. This can be identified by measuring the SoO and pain thresholds after performing the rubber hand illusion (RHI) with an injured rubber hand. The generation of SoO is thought to be caused by multisensory integration of bottom-up factors (vision, proprioceptive, and touch), and by top-down factors, such as the context effect. The appearance is one of the context effects which may become more effective when used simultaneously with other context effects (e.g., verbal suggestion). However, in the RHI, when appearance and other context effects are used simultaneously, the effect is unclear. In this study, we attempted to identify the influence of verbal suggestion on the SoO and heat pain threshold (HPT). As a preliminary step, in Experiment 1, the "normal" rubber hand and "penetrated nail" as injured rubber hand were used to clarify the context effect with appearance alone during RHI (synchronous/asynchronous), which was conducted within-subjects. In Experiment 2, we only used the "penetrated nail" rubber hand to clarify the context effect with verbal suggestion and appearance during RHI. We randomly classified participants into two suggestion groups ("fear" and "no-fear"). The RHI (synchronous/asynchronous) was conducted for each group. In each experiment, the effect of each condition was assessed by subjective measures of SoO, such as questionnaire, and objective measures of SoO, such as proprioceptive drift and electrodermal activity. Following RHI in each condition, HPT was measured. The main finding was that, in the synchronous condition, the "penetrated nail" appearance with "fear" verbal suggestion modulated questionnaire and HPT, but not electrodermal activity. We conclude that the context-included multisensory integration affected the subjective factors because it contains a higher cognitive process by verbal suggestion.

Experience with opioids does not modify the brain network involved in expectations of placebo analgesia

Placebo analgesia (PA) is defined as a psychobiological phenomenon triggered by the information surrounding an analgesic drug instead of its inherent pharmacological properties. PA is hypothesized to be formed through either verbal suggestions or conditioning. The present study aims at disentangling the neural correlates of expectations effects with or without conditioning through prior experience using the model of PA.

We addressed this question by recruiting two groups of individuals holding comparable verbally‐induced expectations regarding morphine analgesia but either (i) with or (ii) without prior experience with opioids. We then contrasted the two groups' neurocognitive response to acute heat‐pain induction following the injection of sham morphine using electroencephalography (EEG). Topographic ERP analyses of the N2 and P2 pain evoked potential components allowed to test the hypothesis that PA involves distinct neural networks when induced by expectations with or without prior experience.

First, we confirmed that the two groups showed corresponding expectations of morphine analgesia (Hedges' g_s < .4 positive control criteria, g_s = .37 observed difference), and that our intervention induced a medium‐sized PA (Hedges' g_av ≥ .5 positive control, g_av = .6 observed PA). We then tested our hypothesis on the recruitment of different PA‐associated brain networks in individuals with versus without prior experience with opioids and found no evidence for a topographic N2 and P2 ERP components difference between the two groups. Our results thus suggest that in the presence of verbally‐induced expectations, modifications in the PA‐associated brain activity by conditioning are either absent or very small.