Effects of Social Support and Placebo Effect on Pain

“Pain is a distressing experience associated with actual or potential tissue damage with sensory, emotional, cognitive and social components” (Williams & Craig, 2016). This definition is proposed in order to emphasize both emotional, cognitive and social aspects in the experience of pain. Besides this definition, there is a distinction between acute and chronic pain. Acute pain is a sudden pain that does not last any longer.

Acute pain has a warning function against noxious stimuli such as thermal or chemical or any adverse sensory insults. Prolonged pain can occur if pain is related to a chronic disease such as rheumatism, and diabetes or if the pain itself develops into a disease. In most cases, a somatic cause is absent or no longer present, causing the pain to lose its warning function. The biopsychosocial model (Gatchel, Peng, Peters, Fuchs, & Turk, 2007) can be used to explain the development of chronic pain. The individual chronic pain is therefore composed of a dynamic interaction of biological factors such as nerve and joint damage or muscle tension which can be explained by the gate control theory and the neuromatrix theory, psychological factors such as anxiety, anger, depression, resilience and social factors such as the family environment and the professional situation.

Another psychological factor that plays an important role in pain perception is pain catastrophizing. If, a high pain catastrophizing value is present, the pain experience is perceived more intensively and vice versa (Kjøgx et al., 2016; Kristiansen et al., 2014). Along with the catastrophizing effects, another immediate psychological factor - social support plays an important role in social factors, as it has a positive effect on the reported pain intensity and coping strategies (Evers, Kraaimaat, Geenen, Jacobs, & Bijlsma, 2003; Klapow et al., 1995; López-Martínez, Esteve-Zarazaga, & Ramírez-Maestre, 2008). Also in the area of acute pain treatment, social support can lead to an attenuated perception of pain (Chalmers, Wolman, Nikodem, Gulmezoglu, & Hofmeyer, 1995; Con, Linden, Thompson, & Ignaszewski, 1999; Hodnett, Gates, Hofmeyr, & Sakala, 2013; Klaus, Kennell, Robertson, & Sosa, 1986).

Social Support

Barrera (1986) distinguishes three concepts of social support: social embeddedness, perceived social support, and enacted social support. Social embeddedness means the connection and social network of other persons. Perceived social support is seen “as the cognitive appraisal of being reliably connected to others.” This concept overlaps very much with Cobb’s definition: social support is an “information leading the subject to believe that he is cared for and loved, esteemed, and a member of a network of mutual obligations” (Cobb, 1976). The social action of giving support is the content of enacted social support. Ditzen and Heinrichs (2007) distinguish furthermore between perceived support with the general expectation to be supported as a cognitive schema of one person and actually received support. Sarason, Pierce, and Sarason (1990) have found that only the perceived social support has a relation to health.

Models for social support

Different models are used to explain the positive effects on health. One is the main-effect model which describes a direct positive effect of support on different health parameters. This effect exists independently of a persons’ stress level. The other model is termed buffering-model, it represents the understanding of prevention or reduction of the negative effects of stress on health resulting from support. Both mechanisms are not mutually exclusive and there is evidence for a positive effect of social support and reduction of negative effects of stress. Therefore the models can exist next to each other, whereby it can be assumed that the processes differ for social integration and functional support (Cohen & Wills, 1985). Social integration is comparable to a persons’ social network that a person has around them. Whereas functional support covers emotional support, informational support, social companionship, and instrumental support.

Effects of social support

The results of different studies on the effects of social support are partly contradictory. Overall social support influences different parts of life in many ways. For instance, in use of alternative medicine perceived support by a friend is positively associated and the use of biologically based therapy is positively associated with perceived partner support. (Honda & Jacobson, 2005). Social support has a positive effect as well on medication adherence, it is suggested that social support influences patient adherence through its impact on health (DiMatteo, 2004). The assumed mode of action is positive beliefs that are related to better adherence (Luszczynska, Sarkar, & Knoll, 2007). Another study showed positive associations between social support and hugs with infection and illness signs (Cohen, Janicki-Deverts, Turner, & Doyle, 2015). Social support is also found in conjunction with decreased morbidity and mortality (Berkman, 1985; House, Landis, Karl, Umberson, Landis, & Umberson, 1988).

Literature also shows instances in interoceptive (physiological) body responses, that the presence of a friend moderates the cardiovascular responses to stress (Edens, Larkin, & Abel, 1991). It was also shown that this effect is greater when the support comes from a friend than from a stranger (Christenfeld et al., 1997).

Social support and secure attachment also have an attenuating effect on anxiety during stress exposure (Ditzen et al., 2008), furthermore the cortisol levels recover faster after stress exposure (Meuwly et al., 2012). Besides, social support also affects the general cortisol response (Kirschbaum, Klauer, Filipp, & Hellhammer, 1995) Physiological systems which are related to stress are beneficially influenced by supportive behavior (Holt-Lunstad, Birmingham, & Light, 2008). A meta-analysis could detect an attenuating effect of social support on reactivity (heart rate, blood pressure, skin conductance, and cortisol) and is thereby suggesting that social support has a health-protective action (Thorsteinsson & James, 1999).

Interactions of social support and pain

The pain stimulus was rated as less threatening in the presence of an observer who could be seen as a safety cue, but the social presence enhanced the facial expression of pain (Vlaeyen et al., 2009). The effect of showing higher levels of pain behavior when a supportive person is around was also in the study of (Gil, Keefe, Crisson, & Van Dalfsen, 1987). The pain report increased as well in women when a friend was present. This effect can be explained by social reinforcement models in chronic pain (McClelland & McCubbin, 2008). But on the contrary fibromyalgia patients reported less pain and thermal pain sensitivity when the significant other was present (Montoya, Larbig, Braun, Preissl, & Birbaumer, 2004). Even viewing a picture of a romantic partner enhances analgesia (Master et al., 2009; Younger, Aron, Parke, Chatterjee, & Mackey, 2010).

The pain ratings of acute pain such as labor pain (Chalmers et al., 1995; Cogan & Spinnato, 1988; Hodnett et al., 2013; Klaus et al., 1986; Niven, 1985) or postoperative pain(Con et al., 1999; Kulik & Mahler, 1989) can be reduced by social support and can also lead to reduced pain medication use. Lepore (1998) represents the importance of the distinction of passive and active support paradigm and distinguishes them in that passive support is explained by the presence of a person which is forbidden to communicate with eye contact or verbal contact. Active support is characterized by the fact that the supporting person makes supporting gestures and comments. In the study by Brown, Sheffield, Leary, and Robinson (2003) the participants in the passive and active support condition reported less pain than in the alone condition and interaction condition. According to a meta-analysis by Che, Cash, Chung, Fitzgerald, and Fitzgibbon (2018) pain ratings like intensity and unpleasantness and pain sensitivity like threshold and tolerance did not differ with the presence of a close accompanying person, but the presence did increase the facial expression.

The meta-analysis applied the standard mean difference (textit{SMD}) using Hedge’s adjusted g. Verbal support of a known person has textit{SMD} = -0.69. In a systematic review Krahé, Springer, Weinman, and Fotopoulou (2013) suggested that positive and structured comments (Brown et al., 2003; Chambers, Craig, & Bennett, 2002; Jackson, 2007; Roberts, Klatzkin, & Mechlin, 2015) are related to decreased pain, which is confirmed by the meta-analysis. The pain reduction may also be caused by the buffering effect of verbal support on the reduced physiological response (textit{SMD} = -0.99 in Roberts et al., (2015)). Especially verbal support can increase intimacy and satisfaction resulting in a less intense pain experience and perceived threat (Corley, Cano, Goubert, Vlaeyen, & Wurm, 2016; Leong, Cano, Wurm, Lumley, & Corley, 2015). This effect can be enhanced be intranasal given oxytocin (Heinrichs, Baumgartner, Kirschbaum, & Ehlert, 2003). Social touch that was given by close others decreased behavioral pain. The study by Che, Cash, Fitzgerald, & Fitzgibbon (2018) found that the analgesic influence of social support passes over the modulation of the threat of pain.

Social touch and oxytocin

Hand-holding as a social touch is a commonly used nonverbal gesture to express support and attachment. Repeated warm touch enhances the oxytocin level and reduces the stress level (Holt-Lunstad et al., 2008), when the relationship is considered positive (Uvnäs-Moberg, Handlin, & Petersson, 2015). Hand-holding of a spouse had a less threat-related neural activation (Coan, Schaefer, & Davidson, 2006) and intranasally given oxytocin appears to increase the buffering effect of social support on the stress response (Heinrichs et al., 2003). Related results showed that the oxytocin level is constant if touch is provided (Henricson, Berglund, Määttä, Ekman, & Segesten, 2008). But the findings in humans on a touch-induced release of oxytocin are heterogeneous (Walker, Light, Løseth, Wessberg, & Olausson, 2016). For example in the study of Ditzen et al. (2007) women who participated with positive physical contact showed lower cortisol levels but no difference in oxytocin levels. Seltzer, Ziegler, and Pollak (2010) suggest that vocalization may be as important as touch in regulation of oxytocin.

Analgesic effect of social touch and oxytocin

The analgesic effect of social touch is powerful and may be mediated by the empathy of the romantic partner (Goldstein, Shamay-Tsoory, Yellinek, & Weissman-Fogel, 2016).

Oxytocin modulates the positive effects of social support and influence pain inhibition (Kreuder et al., 2019).

Placebo

The term placebo comes from Latin and means 'I will please'. In the eighteenth century did William Cullen (Stott, 1987) used the term placebo in the medical context. He administered the placebo to a patient with an incurable disease in order to please or comfort him. At a similar time, placebos were used as control condition in the medical context. In blinded randomized clinical trials (RCTs) it is useful to apply a placebo-controlled trial to prevent test persons knowing the kind of treatment and building expactations (Rief & Glombiewski, 2012). Placebos are still used as controls and the mechanisms of the placebo effect are also being explored. A definition of placebo does not exist, but a more widely used one is of (Shapiro & Shapiro, 2000): “A placebo is any treatment (including drugs, surgery, psychotherapy, and quack therapy) that is used for its ameliorative effect on a symptom or disease but that actually is ineffective or is not specially effective for the condition being treated”. The psychological aspect of the placebo mechanism is not explained in this definition.

The term placebo effect is used in clinical studies and refers to any improvement in the state of health of a person after therapy. In contrast to this, the placebo response is the biological, physiological reaction to the placebo treatment.

Mechanisms

The mechanisms of placebo response include expectations that can be evoked by verbal suggestions, conditioning and social learning. Verbal suggestions including the suggestion that the patient should remember a previous pain relieving event or that he will receive an effective pain medication can lead to a pain relieving reaction. By manipulating the pain intensity given to the patient verbal suggestions can be amplified (Colloca, Klinger, Flor, & Bingel, 2013). In addition, associative learning can also be used in the form of classical conditioning by giving the patient a pain medication in the acquisition phase so that the expectation is formed that the medication leads to a pain reduction and then replaces the medication with a placebo, which then leads to the same reaction in the patient (Medoff & Colloca, 2015). Also social observational learning is method to induce a placebo response. In the study of (Colloca & Benedetti, 2009) has been observed learning compared with conditioning and verbal suggestions. In the group that should learn the placebo response by observational learning, a demonstrater was observed that showed an analgesic effect. The placebo response in this group was similar to the conditioning group and only the verbal suggestion led to a lower response. This effect could also be shown when the demonstration of the analgesic effect was shown on a video, whereas the empathetic concern does not play a role in the video based observation, which had an effect in the live-observation (Hunter, Siess, & Colloca, 2014).

Social and interpersonal factors on placebo analgesia

In the area of social and interpersonal factors of influence on placebo analgesia, research is not very far advanced. The focus so far has been on the patient-provider relationship. A supportive patient-provider relationship can lead in a clinical context to a twice greater reduction in the severity of disease than the placebo response alone (Kaptchuk et al., 2008). One study showed that the placebo response along with positive expectations of the treatment was greater in the group treated by a warm and competent drug provider than in the group treated by a cold and incompetent one (Howe, Goyer, & Crum, 2017). In addition, demonstrating care and empathy of the provider may lead to an increased positive response to a clinical setting and may also activate psychosociobiological adaptations leading to calmness and satisfaction associated with the placebo phenomenon (Blasini, Peiris, Wright, & Colloca, 2018). Also loneliness is considered to influence the placebo response as it influences the therapeutic alliance (Necka & Atlas, 2018). Persons who feel lonely or socially isolated assume that they are being poorly treated (Cacioppo & Hawkley, 2005; Jones, Freemon, & Goswick, 1981; Lamster, Nittel, Rief, Mehl, & Lincoln, 2017), this leads to a difficulties building the therapeutic alliance that is important for the placebo response. Since there is a negative connection between loneliness and social support (Bernardon, Babb, Hakim-Larson, & Gragg, 2011; Kong & You, 2013), one could assume that social support has a positive effect on the placebo response. However, the effects of social support of friends and romantic partners on the placebo response has not yet been investigated.

Personalized medicine

Recent medical research deals with personalized medicine to improve the treatment of diseases. Hood (2006) adopted the approach of 'predictive, personalized, preventive, participatory (P4)' proposed for treatment methods to be improved with an analysis of the genetics of the individuals and also for individuals to get a more active role in deciding on the health. The approach has been further developed, particularly for cancer treatment (Hood & Friend, 2011). Since medical research is mainly concerned with biomarkers for predicting and treating diseases, Gorini and Pravettoni (2011) claim that it is necessary to include the 'psycho-cognitive' factor in a P5 approach. The aim is to adapt treatment methods to personality factors. As personality is linked to health behavior, it should also be collected in particular in prevention (Israel et al., 2015). “Furthermore, experiments in laboratory rodents clearly reveal that certain physiological and neuroendocrine characteristics related to personality are strong indicators for pathology development.” (Boersma, Benthem, van Beek, van Dijk, & Scheurink, 2011)

In research of depression there are studies on the personalization of treatment methods that have assigned patients in the experimental group to the most suitable treatment method with personality predictors and moderators. In contrast to the control group, the effect size of the clinically significant improvement of the group with the personalized treatment was 0.51 - 0.58 greater than that of the control group (DeRubeis et al., 2014; Huibers et al., 2015). Since clinical pain consists of different components and these also interact, it is difficult to find predictors of individual pain, but with the “quantitative systems pharmacology approach” it is possible to identify them (Goulooze et al., 2017). Also in the field of placebo research it is proposed to personalize the placebo treatment and to identify the psychological, neuroendocrine and genetic moderators or predictors (Enck, Bingel, Schedlowski, & Rief, 2013). In alternative medicine such as Ayurveda and traditional chinese and korean medicine, illnesses have long been personalized based on personality types, as each type is attributed special physical characteristics and reception of medical treatment. However, there is no research on how the suggestion “This is your personalized medicine” can increase the placebo response.

Aims of the study

The study was planned in order to investigate effects on the relief of experimentally induced pain. The study aimed to create a novel study design with a combination of social support, administration of placebo and personalized placebo.