Previously titled: Nocebo as a source of bias in the assessment of treatment effect

The term nocebo effect refers to the harmful outcomes that result from people’s negative beliefs, anticipations, or experiences related to the treatment rather than the pharmacological properties of the treatment. These outcomes may include a worsening of symptoms, a lack of expected improvement, or adverse events, and they may occur after the active treatment and the placebo that is supposed to imitate it.  The nocebo effect is always unwanted and may distort estimates of treatment effectiveness and safety; moreover, it may cause discontinuation of therapy or withdrawal from a trial. The nocebo effect may be unintentionally evoked by the explanations given by healthcare professionals during a clinical consultation or consent procedures, or by information from other patients, the media, or the Internet. Moreover, it may be a consequence of previous bad experiences with the treatment, through learning and conditioning, and the conditioning may happen without patients’ conscious awareness. In trial settings, a study design, for example lack of blinding, may introduce bias from the nocebo effect. Unlike the placebo effect, which is usually taken into consideration while interpreting treatment outcomes and controlled for in clinical trials, the nocebo effect is under-recognised by clinical researchers and clinicians. This is worrying, because the nocebo phenomenon is common and may have potentially negative consequences for the results of clinical treatment and trials. It is therefore important that doctors and medical researchers consider any potential nocebo effect while assessing the treatment effect and try to minimise it through careful choice and phrasing of treatment-related information given to patients.


Introduction
Nocebo is often described as placebo's evil twin, and is rarely discussed on its own. This phenomenon is under-recognised in clinical practice and clinical trials, and many patients and healthcare professionals admit that they are not aware of its existence (Berthelot et al., 2001).
The nocebo effect is defined as the adverse effects of an intervention that are not related to its pharmacological or physiological effects. In research settings, it refers to the negative effects of a placebo. In clinical or trial settings, the term is used to describe the negative effects produced by placebos or either harm or lack of efficacy of the active intervention (Benedetti et al., 2007;Hahn, 1997;Häuser et al., 2012). In this paper I will not distinguish between a nocebo effect and the nocebo response and I shall refer to all the clinical/psychological or physiological changes in a patient or a group of patients that are not related to the pharmacological or physiological effect of the treatment as the nocebo effect.

Nocebo effect in clinical practice
The consequences of the nocebo effect in clinical practice are always undesirable. It may make therapeutic interventions more painful, reduce responses to treatment, worsen symptoms or lead to adverse events, in turn causing non-adherence or discontinuation of treatment (Blasini et al., 2017). For example, an injection of an epidural analgesic can be made more painful when patients are warned that it would feel like a bee sting rather than told only that it would create a numbing sensation (Varelmann et al., 2010). Similarly, using the word "pain" rather than "a cool sensation" in a description of a procedure may make this procedure painful (Lang et al., 2005). Also, the efficacy of pharmacologically active substances was greatly reduced when they were given with contradictory information, for example bronchoconstrictors as reducing asthma and bronchodilators as provoking it (Luparello et al., 1970). Information that injection of a powerful opioidergic analgesic was started or stopped increased or decreased its analgesic effects, despite continued delivery of the drug (Bingel et al., 2011). In another study, pain ratings after the suggestion of hyperalgesia were higher than after the suggestion of analgesia, regardless of whether they were accompanied by an application of an analgesic cream or a placebo (Aslaksen et al., 2015).
The nocebo effect may also be responsible for worsening of symptoms or for new symptoms, which are often recorded as adverse effects related to the medication. For example, nebulised saline evoked asthma attacks in patients with asthma if it was given with information that contained an irritant, while the same saline relieved the symptoms if it was presented as an active treatment (Luparello et al., 1968). This is particularly important in the context of clinical trials, as a patient may be asked to record any unusual symptoms and the doctor, who may not know the patient's long-term clinical history, may be more likely to interpret any new symptom as an adverse effect of the medication. Adverse events caused by the nocebo effect are discussed in more detail in the section on nocebo in clinical trials.
In clinical settings, the nocebo effect can be easily evoked through verbal suggestion (Benedetti et al., 2007). These negative effects are usually created unintentionally, by the description of the treatment effects and adverse effects during a clinical consultation or during a consent process (Benedetti et  It may also result from one's bad experiences, through learning and conditioning. The nocebo effect may be caused by dissatisfaction with past treatment, for example due to lack of treatment

Amendments from Version 1
This manuscript has been revised to incorporate the reviewers' comments. In the revised version, the title has been amended, additional subheadings have been added, and some paragraphs have been rearranged to follow the clinic/trial context subdivision, and the possible causes and consequences of the nocebo effect are mentioned. Moreover, the sections on classical and operant conditioning as well as learning by observation have been extended. For more details, please see the responses to the reviewers. The nocebo effect may also be caused by a negative attitude to the treatment or dissatisfaction with either the treatment or the doctor. For example, patients often distrust generic drugs and believe that they are less effective and more harmful than branded drugs (

The nocebo effect in clinical trials
In clinical trials, the nocebo effect manifests as reduced improvement or increased frequency of adverse events, in both the placebo and treatment arms. Patients' withdrawal from a trial due to these adverse events is also considered to be a nocebo effect . The nocebo effect in clinical trials is undesired and may distort the results of the trial; for example, if patients do not improve sufficiently, it may be concluded that the tested treatment is ineffective, or if patients report many adverse events, the conclusions may be that the treatment is harmful and the trial may be terminated early. Moreover, if these adverse events lead to the withdrawal of many participants, the missing data may further complicate interpretation of the results (Mitsikostas  et al., 2011).
The nocebo effect is common, but, unlike the placebo effect, it is rarely discussed in the context of clinical trials, and it may not be considered when interpreting the results of a study. This In clinical trials, as in the clinic, the nocebo effect may be introduced by negative information about the beneficial and adverse effects of the tested treatment that are described in the information letter or during the informed consent process, and not by the pharmacological properties of the treatment (Barsky et al., 2002). This may bias trial outcomes, especially if these outcomes are based on patients' reports. For example, the frequency of reported gastrointestinal adverse events and discontinuation rates due to these adverse events in a trial of aspirin were much lower in a centre that did not include information about possible gastrointestinal bleeds than in two centres that included this information (Cairns et al., 1985; Myers et al., 1987). Moreover, about a quarter of patients taking placebo spontaneously report at least one adverse event, and this figure increases The trial design itself may introduce the nocebo effect and undermine its results. For example, random assignment to different treatment regimens means that patients are not given a choice, which may create a nocebo effect (Bartley et al., 2016). However, blinding of patients and assessors reduces placebo and nocebo bias, because it tends to make the conditions and expectations identical in the active and placebo groups (Collins & MacMahon, 2007). Lack of blinding in one study arm (observational/waiting list group) or all study arms (open-label trials) may distort trial results, as knowledge about the received treatment may affect the incidence of reported adverse events. For example, in a group of patients who knew they were taking atenolol and that erectile dysfunction may a possible adverse effect the incidence of this symptom was 31%; in a group that was informed about the drug but not about the adverse effects, the incidence was 16%, and in the group that was blinded and not told explicitly about this potential effect the incidence was only 3.1%. In the patients who reported this adverse effect, sildenafil or placebo were equally effective in curing it (Silvestri et al., 2003).
Placebo control is useful not only to test whether the active treatment is more effective than placebo but also whether it is truly more harmful than placebo. Without a placebo control, all adverse events may be attributed to the active element of the treatment. For example, in a trial of statins, during the blinded and randomised phase, muscle-related symptoms were reported equally often in the active and placebo arms, but during the unblinded phase they were more frequent in patients taking statins (Ganga et al., 2014;Gupta et al., 2017;Kashani et al., 2006). Moreover, patients with well-documented statin intolerance due to muscle symptoms usually tolerate a statin under doubleblind conditions (Newman & Tobert, 2015). If patients in the control group do not receive placebo but are only followed-up in the trial (so-called observational/waiting list) they may demonstrate deteriorating symptoms or reduced improvement in self-limiting conditions, because they know they are left without any treatment and their treatment expectations are not met. For that reason, a non-interventional arm does not represent the natural history of the disease, because there is a double bias: not only are these patients not blinded, but their treatment expectations are not met, because they are left with no treatment, which leads to the nocebo effect and either worsening of their symptoms or slower recovery.

Recommendations and future directions
Unlike improvement associated with placebo, there are no benefits related to the nocebo effect; it, therefore, has to be minimised by reducing pre-existing negative expectations or by preventing new ones (Tobert & Newman, 2016).
Negative symptoms may not be reported if they are not prompted. It may be beneficial not to inform patients about potential adverse events that may be unrelated to the treatment or be of little clinical importance, such as mild headaches or nausea (Tobert & Newman, 2016). However, it is crucial to warn patients about clinically important or potentially dangerous adverse effects predicted from the pharmacological properties of a drug; for example, warning patients not to drive or operate heavy machinery after medications that cause drowsiness. In clinical trials, it is also very important to record and include in the publication the exact content and phrasing of the information given to the participants, because it may have a substantial effect on the trial results. The nocebo effect can also be prevented by careful phrasing and positive framing of the information given to patients, for example by focusing on chances of improvement, survival, being symptom-free, and of not developing adverse effects (Crichton & Petrie, 2015).
Another way to reduce the nocebo effect is to ask patients about their preconceptions and beliefs regarding a treatment or about previous experiences as they may negatively affect treatment outcomes. If patients' beliefs are negative (if, for example, they think they are intolerant of the prescribed medicine), they will be more likely to report adverse events at follow-up (Barsky et al., 2002), especially when starting new medications (Nestoriuc et al., 2010). Such patients will also be less likely to adhere to treatment (Barsky et al., 2002), and may be more likely to stop taking the medication altogether (Nestoriuc et al., 2010). After a change of medication, patients with negative beliefs tend to report even more adverse events than during therapy with the original drug (Nestoriuc et al., 2010). It is therefore important to change patients' attitudes before changing medications. Moreover, it may be worth asking patients to agree to a re-challenge with a drug they claim they do not tolerate (Tobert & Newman, 2016), as having a choice is associated with better outcomes (Botti & Iyengar, 2004). Finally, it is also important not to leave the patient without treatment, as any appropriately indicated treatment is better than staying on a waiting list (Khan et al., 2012).

Conclusions
The nocebo effect is always negative and unwanted. It can easily be evoked by a careless word or unfortunate phrasing. It can also be learned from one's own bad experiences or by observing others. It may also be caused by classical and operant conditioning, sometimes without patients being consciously aware of it.
Recognising the nocebo effect is important, because it may make a treatment look ineffective or harmful. For example, there may be no improvement or a much smaller improvement than expected, or the medication may seem to be poorly tolerated, with patients reporting many adverse events, which may lead to a change of therapy. However, patients who reported nonspecific complaints after one drug are likely to report even worse symptoms after a change of treatment.
The nocebo effect is also responsible for non-adherence to treatment and for discontinuation. When patients expect to feel worse or not to improve, they treat every negative sensation as being caused by the treatment, and so they do not take the treatment regularly or stop it altogether, which in turn results in a subtherapeutic dose of medication.
Any potential nocebo effect must, therefore, be recognised and minimised in the clinic and in clinical trials.

Data availability
No data is associated with this article.

Grant information
The author declares that no grants were involved in supporting this work.
also important and their role should be discussed, i.e. previous experience (classical conditioning) and observation of other patients/participants of clinical trials (see for example 1). Especially, the role of those two additional sources should be included in the recommendations and conclusions sections of the paper.
Second, through the paper the nocebo effect is discussed mainly in terms of negative expectations, however, it is only one of the explanatory mechanisms of the nocebo effect. Although nocebo effects induced by verbal information and observational learning are usually mediated by expectations, there is growing evidence that the nocebo effect induced by classical conditioning may not always be mediated by expectations (see references 2-8). Thus, I would rather avoid discussing the nocebo effect as the result of sole negative expectancies as well as I would avoid defining it in terms of negative expectations.

Partly
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Pain, memory of pain, placebo and nocebo effects, learning mechanisms of pain and placebo effects I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
how the nocebo effect may be activated. Others include invalidation (mentioned briefly), conditioning (mentioned briefly) and activation of the fight or flight response (not mentioned at all).
Another issue for me is the contexts in which nocebo can be an issue. This article mentions clinical trials, experimental settings such as pain perception, and clinical practice. I would find it easier to navigate the article if there was a clearer differentiation between these very different contexts, perhaps with subheadings.
In relation to clinical practice, the article does not mention the fact that a consultation can make a patient's disease or symptoms worse as a result of nocebo mechanisms.
In the context of clinical trials I think that more attention should be given to the fact that the consent procedure can make symptoms worse, as well as resulting in a reduced response or adverse events.
I found the English a little clumsy in places.

Is the review written in accessible language? Partly
Are the conclusions drawn appropriate in the context of the current research literature? Yes manuscript.
I am grateful to the Reviewer for their comment on the lack of clarity regarding the subdivision of nocebo effect in clinical and trial contexts. In the revised version, additional subheadings have been added, and some paragraphs have been rearranged to follow the clinical/trial context subdivision followed by the possible causes and consequences of the nocebo effect. Hopefully, the new version of the manuscript is less clumsy and sufficiently highlights the role of conditioning, stress, consultation, and consent procedure in generating the nocebo effect.
Competing Interests: I have no competing interests.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com